Jordan v. Maxim Healthcare Services

950 F.3d 724

Opinion

FILED

United States Court of Appeals Tenth Circuit

PUBLISH February 19, 2020 Christopher M. Wolpert

UNITED STATES COURT OF APPEALS Clerk of Court

TENTH CIRCUIT

THERESA JORDAN, individually and on behalf of the Proposed Colorado Rule 23 Class,

Plaintiff - Appellee, No. 18-1290

v.

MAXIM HEALTHCARE SERVICES, INC.,

Defendant - Appellant.

Appeal from the United States District Court for the District of Colorado (D.C. No. 1:15-CV-01372-KMT)

David B. Salmons (Lincoln O. Bisbee, Matthew J. Sharbaugh and Michael E. Kenneally, with him on the briefs), Morgan, Lewis & Bockius LLP, Washington, D.C., for Defendant-Appellant.

Robert E. DeRose, Barkan Meizlish Handelman Goodin DeRose Wentz, LLP, Columbus, Ohio (Jason J. Thompson, Sommers Schwartz, P.C., Southfield, Michigan; Timothy J. Becker and David H. Grounds, Johnson Becker, PLLC, St. Paul, Minnesota, with him on the brief), for Plaintiff-Appellee.

Before BRISCOE, HOLMES, and McHUGH, Circuit Judges.

HOLMES, Circuit Judge.

Under Colorado law, employers generally must pay all employees time-and-

a-half wages for overtime hours. That said, the law carves out several express exemptions from this requirement. Specifically, employers need not pay overtime wages to “companions, casual babysitters, and domestic employees employed by households or family members to perform duties in private residences.” 7 C OLO . C ODE R EGS . § 1103-1:5 (2019) (the “companionship exemption”). The question before us is whether “companions” working for third-party employers—rather than for households or family members—fall within the companionship exemption. We hold that they do. Accordingly, exercising jurisdiction under 28 U.S.C. § 1291, we reverse the district court’s judgment concluding otherwise.

I

Maxim Healthcare Services, Inc. (“Maxim”) is a for-profit staffing company that provides customers with in-home care. Theresa Jordan worked for Maxim in Colorado as a home health-care worker. The parties do not dispute that in that capacity, Ms. Jordan and other similarly situated Maxim employees were “companions” under Colorado law, and Maxim concedes that it did not pay Ms. Jordan or its other “companions” time-and-a-half wages for overtime hours from

3

2012 through 2015. 1 Ms. Jordan, on behalf of a class of Maxim companions, argues that this failure violated Colorado law.

A

Before turning to Colorado law, we offer a primer on federal wage-and-

hour law, which undergirds our analysis of the state-law issues before us. The lynchpin federal law in this case is the Fair Labor Standards Act (the “FLSA”), 29 U.S.C. § 201 et seq., which mandates that employers in interstate commerce pay employees one-and-one-half times their regular hourly rate for all overtime hours worked, id. § 207(a)(1). This overtime rate generally applies to “any employee in domestic service.” Id. § 207(l). At the same time, the FLSA contains exemptions for certain classes of domestic-service employees, thereby relieving their employers from the overtime-pay requirement. Of particular relevance here is the statute’s exemption for domestic-service employees who “provide companionship services for individuals who . . . are unable to care for themselves.” Id. § 213(a)(15).

1 Ms. Jordan initially sought relief dating back to May 27, 2012, on the view that the Colorado Wage Act’s three-year limitations period for “willful violation[s]” was applicable to her claims. C OLO R EV . S TAT . § 8-4-122 (2015). But the district court ultimately granted summary judgment in Maxim’s favor on that issue (an issue that Ms. Jordan does not challenge on appeal), and as such, the period in the final judgment from which Ms. Jordan appeals—and thus the operative period for our purposes—spans from May 27, 2013 (i.e., not 2012) through October 12, 2015.

4

The U.S. Department of Labor, which is charged with implementing the FLSA, has promulgated regulations explicating this exemption. One such regulation, which was in effect during most of the years at issue here, defined “companionship services” as “those services which provide fellowship, care, and protection for a person who . . . cannot care for his or her own needs.” 40 Fed. Reg. 7404, 7405 (Feb. 20, 1975) (language formerly codified at 29 C.F.R. pt. 552, subpt. A, § 552.6). This regulation also specified that employees who “provid[e] companionship services . . . and who are employed by an employer or agency other than the family or household using their services[] are exempt from the [Fair Labor Standard] Act’s . . . overtime pay requirements.” 2 Id. at 7407 (language formerly codified at 29 C.F.R. pt. 552, subpt. B, § 552.109(a)) (emphasis added). In other words, during most of the relevant period, federal law explicitly exempted from the FLSA’s overtime-pay requirement those companions who were employed by third-party employers.

2 As noted supra, this regulation was in effect during most of the relevant period. It was not until January 1, 2015 that the Department of Labor amended its regulations to specify that “[t]hird party employers of employees engaged in companionship services . . . may not avail themselves of the . . . overtime exemption provided by [the FLSA].” 29 C.F.R. § 552.109(a) (effective Jan. 1, 2015) (emphasis added).

5

B

Like the FLSA, Colorado statute provides for overtime pay “at a rate of one and one-half times the regular rate of pay.” C OLO . R EV . S TAT . § 8-6-111(4) (2019). 3 To flesh out this provision, the Colorado General Assembly delegated to the director of the Division of Labor (the “Division”) the authority to prescribe the “conditions and rules” governing overtime compensation. Id. The Division has done so by promulgating each year a new iteration of the Colorado Minimum Wage Order (the “Wage Order”), 4 which regulates, inter alia, wages and hours for certain employers and employees in the state. The only statutory limitation on this authority is that all wage orders “shall apply equally to all employers in [an] industry or occupation.” Id.

3 The 2019 version of the pertinent Colorado statute is identical in all relevant respects to the iterations of that statute that had been in effect from 2013 to 2015, i.e., the years at issue here.

4 As noted supra, the Division issues a new Colorado Minimum Wage Order every year, with each newly issued version superseding all prior versions. The versions in effect during the period at issue in this case were, respectively, Colorado Minimum Wage Order Number 29 (effective in 2013), 30 (effective in 2014), and 31 (effective in 2015). These three versions were identical to one another in all relevant respects—namely, the overtime-pay requirement and companionship exemption—and in fact, the salient language contained therein remains in force to this day. Accordingly, in lieu of citing throughout this opinion to three discrete, long-superseded versions of the Colorado Minimum Wage Order, for the sake of simplicity and ease of reference, all citations and references to the “Wage Order” hereinafter correspond to the 2019 version.

6

This appeal centers around the Wage Order’s overtime-pay requirement and the scope of a certain exemption to that requirement. In particular, section four of the Wage Order prescribes a general rule that all employees working in certain industries 5 must be paid the statutorily set time-and-a-half pay rate for overtime hours. 7 C OLO . C ODE R EGS . § 1103-1:4. Section five then identifies various exemptions to this requirement, including, as referenced supra, the companionship exemption. That exemption provides that “companions, casual babysitters, and domestic employees employed by households or family members to perform duties in private residences” are “exempt from all provisions of [the] Wage Order.” Id. § 1103-1:5. The gravamen of the parties’ dispute is the phrase, “employed by households or family members to perform duties in private residences,” which we frequently refer to herein in shorthand form as the “household modifier.”

The Wage Order is short on details. Significantly, it is silent as to whether third-party employers such as Maxim must pay overtime wages to companions.

5 Those industries are (1) retail and service, (2) commercial support service, (3) food and beverage, and (4) health and medical. 7 C OLO . C ODE R EGS . § 1103-1:1. Because Maxim does not contest that companions fall within one of these enumerated industries, we assume without deciding that the Wage Order’s overtime-pay requirement applies to companions in the first instance. See United States v. Bowline, 917 F.3d 1227, 1231 (10th Cir. 2019) (“[W]hen a party omits an argument from its opening brief, an appellate court has no obligation to consider that argument.”), petition for cert. docketed, No. 19-5563 (U.S. Aug. 13, 2019).

7

Moreover, it does not define such key terms as “companions,” “casual babysitters,” and “domestic employees.” See id. (defining other terms but not those); id. § 1103-1:2 (same).

That said, the Wage Order does not exist in a vacuum, but rather within the broader regulatory scheme within which it was promulgated. To begin, the Division has issued an Advisory Bulletins and Resource Guide (the “Bulletin”) “for general advisory, clarification, and explanatory purposes,” which includes a “Keyword Index” addressing some of the Wage Order’s terms that are central to our analysis. Aplt.’s App. at 263 (Excerpt of Advisory Bulletins & Resource Guide, dated Aug. 2005). The Bulletin, for instance, defines “companion services” as “services which provide fellowship, care and protection for a person, who due to advanced age or physical or mental conditions cannot care for his or her own needs.” Id. at 277. And without qualification, that glossary entry also states the following: “[c]ompanions are exempt from all provisions of [the] Wage Order.” Id. The Bulletin likewise includes an entry for “[c]asual babysitt[ing],” which it defines as “work performed on an irregular or intermittent basis and not performed by an individual whose full time work is babysitting.” Id. at 278. And like companions, “[c]asual babysitters are exempt from all provisions of [the] Wage Order,” according to the Bulletin. Id.

8

Through opinion letters, the Division has provided further context on the companionship exemption. For example, in response to Maxim’s inquiry to the Division in 2006 “concerning the companionship exemption and third party employers,” the Division issued a letter detailing its “position” on the exemption as follows: “the practice of applying the companionship exemption in situations involving third party employers is acceptable under . . . [the] Wage Order.” Id. at 192–93 (Colo. Dep’t of Labor & Emp’t, Div. of Labor, Op. Letter (dated Aug. 3, 2006)). In so advising, the Division explained that “the treatment and interpretation of the companions exemption in the Wage Order was intended to mirror” the federal scheme, which in turn explicitly exempted companions employed by third-party employers from the overtime-pay requirement. Id. at 192. Six years later, the Division sent Maxim another opinion letter reaffirming this interpretation. See id. at 187–90 (Colo. Dep’t of Labor & Emp’t, Div. of Labor, Op. Letter (dated June 14, 2012)).

The Division has followed this interpretation of the Wage Order in practice, too. In response to a claim filed with the Division by a Maxim employee in 2012, the Division reiterated that Colorado’s companionship exemption was “intended to mimic federal law.” Id. at 541 (Division’s Resp. to Maxim Emp.’s 2012 Claim, dated Aug. 28, 2012). Two years later, another Maxim employee filed a claim with the Division, in support of which that employee argued that the

9

companionship exemption applied to only those companions “employed directly by households,” not to those employed by third-party employers such as Maxim. Id. at 544 (Maxim Emp.’s Claim, dated June 24, 2014). The Division rejected that argument and concluded that it lacked “the statutory authority to pursue this matter further,” once again noting its “position” was that “the treatment and interpretation of the companions exemption in the Wage Order is intended to mirror . . . federal law.” Id. at 546 (Division’s Resp. to Maxim Emp.’s 2014 Claim, dated July 23, 2014).

C

Against this statutory and regulatory backdrop, Ms. Jordan filed this putative class action in state court in May 2015. She alleged that Maxim violated Colorado law by refusing to pay her and other Maxim companions overtime wages as required by section 4 of the Wage Order. Maxim removed the suit to federal court.

Once in federal court, Ms. Jordan moved for partial summary judgment as to Maxim’s liability. In her motion, she argued that under the plain language of the companionship exemption, only those companions who were employed by households or family members were exempt from the Wage Order’s overtime-pay requirement. She also pointed out that neither party disputed that Maxim was a third-party employer, that Ms. Jordan and the other class members were

10

“companions” under Colorado law, and that Maxim had not paid Ms. Jordan or its other companions overtime wages from 2012 through 2015. Thus, Ms. Jordan reasoned, Maxim had improperly relied on the companionship exemption in denying Ms. Jordan and the other Maxim companions overtime wages, thereby rendering Maxim liable for all unpaid overtime for the relevant period and entitling Ms. Jordan to summary judgment against Maxim.

Maxim cross-moved for summary judgment. It argued that the household modifier (i.e., the phrase, “employed by households or family members to perform duties in private residences”) applied only to “domestic employees” and not to “companions” or “casual babysitters.” In other words, according to Maxim, the companionship exemption extended to all companions—including those employed by third-party employers. On that reading, Maxim concluded that, as a third-party employer, it had no duty to pay Ms. Jordan or the class of Maxim companions overtime wages.

The district court sided with Ms. Jordan, adopting in all relevant respects her reading of the companionship exemption. It reasoned that under its plain language, the exemption applied only to companions who were employed by households or family members, and thus did not apply to employees working for third-party employers such as Maxim. For that reason, the court granted Ms.

11

Jordan’s summary-judgment motion in relevant part, and entered final judgment against Maxim and in favor of Ms. Jordan.

Maxim now appeals from that judgment. It presents one question for our review: Did the district court err in concluding that third-party employers are precluded from invoking the companionship exemption? 6 Put differently, does the companionship exemption apply to companions working for third-party employers, and not just to those working for households or family members?

II

In answering the foregoing question, we review the district court’s order granting Ms. Jordan summary judgment de novo. See, e.g., Birch v. Polaris Indus., Inc., 812 F.3d 1238, 1251 (10th Cir. 2015). And we apply “the same legal standards the district court applied under Federal Rule of Civil Procedure 56(a).” Helget v. City of Hays, 844 F.3d 1216, 1221 (10th Cir. 2017). To that end, we must draw all reasonable inferences and resolve all factual disputes in favor of the non-moving party. See, e.g., Yousuf v. Cohlmia, 741 F.3d 31, 37 (10th Cir. 2014). We will affirm a grant of summary judgment “if the movant shows that

6 Maxim challenges the final judgment on some alternative grounds.

But because we agree with Maxim’s argument that the district court erred in holding that the companionship exemption applies only to those companions employed by households or family members (and thus not to those employed by third-party employers), we do not reach, and thus express no opinion on, these alternative arguments. See Valley Forge Ins. Co. v. Health Care Mgmt. Partners, Ltd., 616 F.3d 1086, 1094 (10th Cir. 2010) (explaining that we typically answer “only the questions we must, not those we can”).

12

there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” F ED . R. C IV . P. 56(a).

We also review the district court’s interpretation of state law de novo. See, e.g., Phila. Indem. Ins. Co. v. Lexington Ins. Co., 845 F.3d 1330, 1336–37 (10th Cir. 2017). If the state’s highest court has not decided an issue, “our task is to predict how it would rule.” United States v. Badger, 818 F.3d 563, 568 (10th Cir. 2016). To guide our prediction, we may consult persuasive state authority, such as “dictum by [the state’s] highest court” and “precedential decisions by a state’s intermediate appellate courts.” Id. at 569. And ordinarily, we are bound by “our own precedent interpreting a state’s law.” Id.; accord Kokins v. Teleflex, Inc., 621 F.3d 1290, 1295 (10th Cir. 2010).

III

We now turn to the question at hand: Do “companions” employed by third-

party employers—as opposed to by households or family members—fall within the companionship exemption of the Wage Order?

Maxim answers that question in the affirmative. It believes the companionship exemption is best read as including companions employed by third-party employers. The district court’s contrary reading, Maxim contends, creates “glaring redundancy” and “conflicts with the enabling statute for the Wage Order[].” Aplt.’s Opening Br. at 21. Maxim also points out that its

13

interpretation comports “with the Division’s longstanding and consistent interpretation” of the Wage Order. Id. For those reasons, Maxim asks us to reverse the district court’s judgment.

Ms. Jordan disagrees. In her view, the companionship exemption is unambiguous, as “the plain language makes it clear that the [companionship exemption] does not extend to third party employers.” Aplee.’s Resp. Br. at 16–17; see Oral Arg. at 19:15–20:30 (reiterating her argument that the exemption is unambiguous). She further argues that because of the exemption’s lack of ambiguity, it is of no moment that the Division has consistently adopted a contrary interpretation of the exemption.

We agree with Maxim. That is to say, we hold that the companionship exemption applies to all companions—including those employed by third-party employers. As a result, Colorado law did not require Maxim to pay Ms. Jordan or the class of Maxim companions overtime wages. We therefore reverse the final judgment awarding Ms. Jordan and the companion class damages and prejudgment interest.

We explain our holding in four steps. First, we summarize the Colorado interpretive rules that govern our analysis. Second, applying these interpretive rules here we determine that, although the companionship exemption is ambiguous, the ordinary and particular meanings of the relevant terms suggest

14

that the Division intended the exemption to apply to all companions. Third, we explain how Colorado’s canons of construction reinforce that suggestion. Fourth, without deciding if deference applies to the Division’s longstanding and consistent interpretation of the companionship exemption, we discuss why the Division’s interpretation provides persuasive support for our holding.

A

Colorado law governs “[o]ur interpretation and application of the Wage Order” and, more specifically, of the companionship exemption. Deherrera v. Decker Truck Line, Inc., 820 F.3d 1147, 1160 (10th Cir. 2016). And under Colorado law, the rules for interpreting a regulation are the same “basic rules of interpretation which pertain to the construction of a statute.” Regular Route Common Carrier Conference of Colo. Motor Carriers Ass’n v. Pub. Utils. Comm’n, 761 P.2d 737, 745 (Colo. 1988). To that end, we interpret the companionship exemption according to Colorado’s ordinary rules of statutory interpretation. See, e.g., Deherrera, 820 F.3d at 1160–62 (using those interpretive rules to interpret the Wage Order).

“The primary goal of interpretation is to ‘give effect to the intent of the enacting body.’” United States v. Richter, 796 F.3d 1173, 1185 (10th Cir. 2015) (quoting Benuishis v. Indus. Claim Appeals Office, 195 P.3d 1142, 1145 (Colo. App. 2008)). To divine this intent, Colorado courts “look to the plain meaning of

15

the [regulatory] language and consider it within the context of the [regulation] as a whole.” Denver Post Corp. v. Ritter, 255 P.3d 1083, 1088 (Colo. 2011). This exercise ordinarily entails reading words and phrases “according to the rules of grammar and common usage.” C OLO . R EV . S TAT . § 2-4-101 (2019). Words or phrases that “have acquired a technical or particular meaning,” however, are read “accordingly.” Id.

If the enacting body’s intent is evident from the text and context, that meaning controls. See, e.g., Deherrera, 820 F.3d at 1161; Specialty Rests. Corp. v. Nelson, 231 P.3d 393, 397 (Colo. 2010). Put another way, when a regulation has only one plausible meaning as divined from its text and context, we apply that meaning, and our analysis ends there: we do not peer beyond the regulation’s text and context in search of other interpretive clues. See Deherrera, 820 F.3d at 1161; Rags Over the Ark. River, Inc. v. Colo. Parks & Wildlife Bd., 360 P.3d 186, 192 (Colo. App. 2015) (“W[hen] constru[ing] agency regulations[,] . . . we look first to the regulation’s plain language. If the plain language is unambiguous, we need not resort to other canons of construction.” (citations omitted)). But if—and only if—a regulation is ambiguous (i.e., “susceptible to multiple interpretations”) do we then “look to other aids in construction” to ascertain the enacting body’s intent. Specialty, 231 P.3d at 397; see Rags, 360 P.3d at 192; see also City & Cty. of Denver v. Expedia, Inc., 405 P.3d 1128, 1132

16

(Colo. 2017) (holding that if a statute is “clear and unambiguous, . . . it must simply be applied as written,” but if it is ambiguous, “interpretive aids, or canons of construction,” are “available to help determine which . . . reasonable interpretation[] . . . embodies the legislative intent” (emphasis added)).

Finally, with respect to Maxim’s burden of proving that Ms. Jordan and the companion class fell within the companionship exemption’s terms, Maxim was required to prove, according to the district court, “that a particular employee ‘plainly and unmistakably’ qualifies for an overtime exemption.” Aplt.’s App. at 286 (Order Den. Mot. to Dismiss, dated Mar. 17, 2016) (emphasis added) (quoting Kennett v. Bayada Home Health Care, Inc., 135 F. Supp. 3d 1232, 1238 (D. Colo. 2015), appeal docketed, No. 19-1004 (10th Cir. argued Nov. 19, 2019)). In employing this plainly-and-unmistakably standard, the district court relied on language from, inter alia, Chase v. Farmers Insurance Exchange, a Colorado state case in which the court, like here, contended with the question of whether certain employees fell under an exemption to a certain wage order. See 129 P.3d 1011, 1014–15 (Colo. App. 2004). In resolving that question, the court invoked the plainly-and-unmistakably standard, reasoning that this standard was applicable there because “[u]nder the FLSA, the employer bears the burden of demonstrating that its employee ‘plainly and unmistakably’ qualifies for an exemption,” and the court “perceive[d] no reason why that same burden should not also be placed on

17

an employer seeking to establish that an employee falls within an exemption under Colorado law.” Id. (quoting Daniels v. Caleel+Hayden, L.L.C., 72 P.3d 466, 468 (Colo. App. 2003)). In other words, the Chase court effectively incorporated the plainly-and-unmistakably standard used to analyze FLSA exemptions under federal law into cases that, like here, concern wage-order exemptions under Colorado law.

Despite Chase’s dictates, Maxim argues on appeal that it need not meet this elevated plainly-and-unmistakably standard. For support, it invokes Encino Motorcars, LLC v. Navarro, 138 S. Ct. 1134 (2018), in which the Supreme Court rejected a hitherto-employed principle by federal courts that FLSA exemptions must be narrowly interpreted, on the view that courts “have no license to give [an] exemption anything but a fair reading.” Id. at 1142. By rejecting the principle that FLSA exemptions must be narrowly interpreted, according to Maxim, Encino renounced the plainly-and-unmistakably standard. See id. at 1148 n.7 (Ginsburg, J., dissenting) (suggesting that Encino toppled the “longstanding principle” that FLSA exemptions apply only “to those [cases] plainly and unmistakably within their terms and spirit’” (alteration in original) (quoting Arnold v. Ben Kanowsky, Inc., 361 U.S. 388, 392 (1960)). Thus, by Maxim’s logic, because Chase had concluded that the Colorado standard for invoking a wage-order exemption tracks the federal standard under the FLSA, and because Encino rejected the plainly-

18

and-unmistakably standard, the plainly-and-unmistakably standard no longer applies to the interpretation of wage-order exemptions under Colorado law post- Encino, and the district court erred in concluding to the contrary here.

In her appeal brief, Ms. Jordan does not contest Maxim’s argument that the plainly-and-unmistakably standard no longer applies post-Encino. See Aplee.’s Resp. Br. at 14 (arguing only that Maxim “severely overstate[s]” the district court’s use of the plainly-and-unmistakably standard, and that the district court had in actuality based its interpretation of the companionship exemption on “the ‘basic rules of interpretation’” (quoting Richter, 796 F.3d at 1185)). And, at oral argument, Ms. Jordan conceded that she had not preserved this argument. See Oral Arg. at 17:20–18:25. Given this concession, we assume without deciding that Maxim need not show that Ms. Jordan and the Maxim companion class plainly and unmistakably fall within the companionship exemption’s terms. Cf. United States v. Iley, 914 F.3d 1274, 1279–80 (10th Cir. 2019) (assuming that de novo review applied, despite the fact that the Tenth Circuit had yet to address whether this standard of review applied in such a case, “because the parties present no genuine controversy regarding it”). Rather, for our purposes, Maxim only needs to show that it may invoke the exemption under a “fair reading.”

19

B

Applying Colorado’s interpretive rules here, we first look to whether the enacting body’s intent is clear from the regulation’s text and context. The enacting body here is the Division, and the relevant text is the companionship exemption. As we explain below, the companionship exemption is ambiguous. And although we believe that the ordinary and particular meanings of the relevant terms suggest that the Division intended the companionship exemption to apply to companions employed by third-party employers (such as Maxim), that suggestion is not so strong as to render the exemption unambiguous.

1

To determine whether the Division’s intent regarding the companionship exemption is clear, we begin by consulting the relevant text. That text—viz., the companionship exemption—provides that “the following employees or occupations . . . are exempt from all provisions of [the Wage Order, including the overtime-pay requirement]: . . . companions, casual babysitters, and domestic employees employed by households or family members to perform duties in private residences.” 7 C OLO . C ODE R EGS . § 1103-1:5.

The parties advance two different readings of this text, with the crux of their dispute concerning the proper construction of the household modifier (i.e., the phrase “employed by households or family members to perform duties in

20

private residences”). Maxim, on the one hand, argues that the household modifier modifies the term “domestic employees” only; that is, it does not also modify the preceding terms “companions” and “casual babysitters.” On this reading, all companions fall under the companionship exemption, regardless of their employer’s identity. Ms. Jordan, by contrast, maintains that the household modifier not only modifies “domestic employees,” but that it also modifies “companions” and “casual babysitters.” Read that way, the companionship exemption applies to only those companions “employed by households or family members to perform duties in private residences” and would thus not apply to Ms. Jordan or to Maxim’s other companion employees. Maxim’s and Ms. Jordan’s readings are both plausible. We have no reasoned way to pick between them, absent an examination of the language’s context, to which we turn infra. Cf. Brunson v. Colo. Cab Co., 433 P.3d 93, 97 (Colo. App. 2018) (concluding that the term “interstate drivers” in the Wage Order was ambiguous), cert. granted then dismissed, 2019 WL 578344 (Colo. Jan. 29, 2019).

To be sure, the district court proclaimed to the contrary that the companionship exemption had only one “grammatically sound” reading. Aplt.’s App. at 289 (quoting Kennett, 135 F. Supp. 3d at 1241). It determined that Maxim’s interpretation was unsound because it “fails to account for its grammatical structure”—i.e., the commas separating the terms “companions” and

21

“casual babysitters,” and the “and” that appears immediately before “domestic employees”—and because such an interpretation would render the “and” preceding “domestic employees” “either superfluous or nonsensical.” Id. at 288 (quoting Kennett, 135 F. Supp. 3d at 1241). The only phrasing of the companionship exemption that would have rendered the exemption applicable to all companions, according to the district court, would have been the following: “[o]ther exemptions are: companions and casual babysitters, and domestic employees employed by households or family members . . . .” Id. at 288–89 (quoting Kennett, 135 F. Supp. 3d at 1241). But the Division did not draft the companionship exemption that way. So, the district court concluded that the only sound reading of the companionship exemption was that it applied to only those companions employed by households and family members.

On appeal, Ms. Jordan echoes the district court, and argues, more specifically, that the companionship exemption is unambiguous. But beyond insisting that the exemption’s plain meaning is obvious, she declines to explain how or why that is so. And other than to say that the district court “determined that no such ambiguity existed,” Ms. Jordan does not defend the particulars of the district court’s supposed declaration to this effect. Aplee.’s Resp. Br. at 14, 16–17.

22

We reject the reading adopted by the district court and Ms. Jordan. To be sure, commas and conjunctions are important. Indeed, where a drafter places a comma can sometimes altogether change a sentence’s meaning. See, e.g., A NTONIN S CALIA & B RYAN A. G ARNER , R EADING L AW : T HE I NTERPRETATION OF L EGAL T EXTS 161–63 (2012). Likewise, a statute’s use of “and” rather than “or” can sometimes alter its meaning. See, e.g., Colo. Med. Bd. v. Office of Admin. Courts, 333 P.3d 70, 74 (Colo. 2014) (concluding that an exemption using “and” was broader than a similar one using “or”). But more often, commas and conjunctions do not imbue a sentence with only one meaning. See, e.g., Expedia, Inc., 405 P.3d at 1137 (noting that commas setting off a conjunctive phrase suggested one meaning, but ultimately adopting another reading because meaning “is ultimately dictated by context”); Clyncke v. Waneka, 157 P.3d 1072, 1079 (Colo. 2007) (Coats, J., concurring in the judgment) (“The word ‘and’ is notoriously ambiguous and has been recognized as such since time immemorial.”). This is one such case.

To see the district court’s error, consider a sign in an apartment complex that says, “No pets allowed. Exceptions include: cats, small dogs, and animals owned by visually impaired persons.” There is no rule of grammar dictating that only cats that are owned by the visually impaired are allowed in the complex. In fact, despite the commas and the placement of “and,” the more natural reading is

23

that the complex permits all cats—regardless of their owners’ visual ability. And just as our fictional sign may be properly read as allowing all cats, the companionship exemption may be properly read as exempting all companions—including those employed by third-party employers. Simply put, the companionship exemption is ambiguous, and the district court erred in concluding otherwise.

2

In view of this ambiguity, we must take a deeper dive into the meaning of the terms “companions,” “casual babysitters,” and “domestic employees.” And this inquiry hints that the companionship exemption applies to all companions. To begin, because neither the General Assembly nor the Division has defined these terms in, respectively, Colorado’s wage-and-hour statutes or the Wage Order, we “look first to the commonly accepted meaning of th[ose] term[s].” Binkley v. People, 716 P.2d 1111, 1113 (Colo. 1986). “When determining the plain and ordinary meaning of words,” the Colorado Supreme Court often looks to “definitions in a recognized dictionary.” Renfrand v. N.Y. Life Ins. Co., 419 P.3d 576, 580 (Colo. 2018). Such dictionaries include Black’s Law Dictionary (“Black’s”), the Oxford English Dictionary (“Oxford English”), the American Heritage Dictionary of the English Language (“American Heritage”), and

24

Merriam-Webster Online Dictionary (“Merriam-Webster”). 7 By consulting these recognized dictionaries, we can identify the commonly accepted meanings of the terms “companions,” “casual babysitters,” and “domestic employees.”

We start with “companions.” At its most basic, the word “companion”

refers to “[a] person employed to assist, live with, or travel with another.” Companion, A MERICAN H ERITAGE 384 (3d ed. 1992). This assistance ordinarily entails “help[ing] with personal matters such as bathing and dressing.” Companionship Services, B LACK ’ S (11th ed. 2019). Due to the intimacy of this assistance, and by virtue of living with the person served, a companion is often “treated more as a friend than an employee.” Companion, O XFORD E NGLISH (3d ed. 2014), www.oed.com/view/Entry/37402. A person need not be employed directly by the individual he or she assists to qualify as a companion, as that term is ordinarily understood. In fact, one might naturally say, “[w]e hired a companion for our elderly mother.” Companion, M ERRIAM -WEBSTER . COM , www.merriam-webster.com/dictionary/companion (last visited Feb. 16, 2020) (providing example of accepted use of “companion” in a sentence). In sum, the

7 See, e.g., Cowen v. People, 431 P.3d 215, 219 (Colo. 2018)

(consulting Black’s and Merriam-Webster’s for a term’s “plain meaning”); Sooper Credit Union v. Sholar Grp. Architects, P.C., 113 P.3d 768, 772 (Colo. 2005) (consulting American Heritage); Copeland v. MBNA Am. Bank, N.A., 907 P.2d 87, 91 (Colo. 1995) (consulting Oxford English); Bringle Family Tr. v. Bd. of Cty. Comm’rs, --- P.3d ----, 2018 WL 2054916, at *4 (Colo. App. May 3, 2018) (consulting Black’s, Merriam-Webster, and Oxford English for a “term’s generally accepted meaning”).

25

ordinary meaning of “companion” is someone employed to help another individual with personal matters and who, through this employment, has a close personal connection with that other individual.

We next turn to the ordinary meaning of “casual babysitters.” The various above-referenced dictionaries agree that “casual” in the employment context means “occurring without regularity.” 8 Casual, B LACK ’ S . These dictionaries likewise concur on the meaning of “babysitter,” defining the term as “[a] person engaged to care for one or more children when the parents or guardians are not at home.” 9 Babysitter, A MERICAN H ERITAGE , supra, at 133. So a “casual babysitter” is a person who irregularly or occasionally is hired to care for children while their guardians are not home.

As for the ordinary meaning of “domestic employees,” none of the referenced dictionaries has a dedicated entry for the term, but we can still

8 See also Casual, A MERICAN H ERITAGE , supra, at 299 (“Occurring at irregular or infrequent intervals[.]”); Casual, M ERRIAM -WEBSTER . COM , www.merriam-webster.com/dictionary/casual (last visited Feb. 16, 2020) (“[O]ccurring without regularity.”); Casual, O XFORD E NGLISH D ICTIONARY , https://www.oed.com/view/Entry/28629?redirectedFrom=casual#eid (“In such phrases as casual labourer, one who does casual or occasional jobs, but has no fixed employment[.]”).

9 See also Babysit, M ERRIAM -WEBSTER . COM , https://www.merriam-webster.com/dictionary/babysit (last visited Feb. 16, 2020) (“[T]o care for children usually during a short absence of the parents.”); Babysitter, O XFORD E NGLISH , www.oed.com/view/Entry/276129? (“A person engaged to look after a child or children while the parents or guardians are out.”).

26

ascertain its commonly accepted meaning. The word “domestic,” which in this context is an adjective, means “[o]f or relating to the family or household.” Domestic, A MERICAN H ERITAGE , supra, at 550. This adjective modifies the noun “employee,” which the Wage Order defines as “any person . . . performing labor or services for the benefit of an employer” and who is not “primarily free from control and direction in the performance of the service.” 7 C OLO . C ODE R EGS . § 1103-1:2. As a result, the ordinary meaning of “domestic employee” is a person employed to perform services relating to the family or household.

An inference arises from these ordinary definitions. Companions and casual babysitters have a close personal connection with the individuals receiving their services. Companions have such a connection with the individual whom they assist with personal matters. Casual babysitters have such a connection with the children they look after. And domestic employees “employed by households or family members to perform duties in private residences” have a similar connection. A maid, for instance, who is hired by a family to come into their home and clean twice a month, has a close personal connection to that family (albeit, typically to a lesser extent than a companion), as evinced by the family’s opening the doors of their home to her, allowing her access to the personal and intimate spaces of their home to clean, and placing their trust in her not to steal or otherwise engage in misconduct in such spaces while she works. The common

27

thread running through these three employee categories is a close personal connection—a commonality that indicates that the Division intended the companionship exemption to apply to household workers who have such a connection with those receiving their services.

This inference begets another. As mentioned, all “companions” have the requisite close personal connection to those whom they serve. This is true even of those companions who are employed by third-party employers: such companions have a close personal connection with the individual with whom they may live and whom they assist with bathing, dressing, or other personal matters. And so the household modifier is unnecessary to effect the Division’s intent as it applies to companions. In fact, applying the modifier to companions would frustrate that intent, as it would exclude some companions—namely, those employed by third-party employers—to whom the Division intended the companionship exemption to apply. The same is not true for domestic employees, however. A maid employed to clean a home by a third-party housekeeping service, as opposed to one employed directly by a family to perform these services, ordinarily will not have a close personal connection with those individuals receiving her services—if any connection at all. After all, those individuals have not personally placed their trust in the maid not to steal or otherwise engage in misconduct in the private and intimate spaces of their home

28

while cleaning; rather, those individuals are holding the third-party employer responsible for vetting its employees and ensuring that they are fit to enter and clean the private homes of strangers. Thus, the application of the household modifier to the term “domestic employees” ensures that the companionship exemption extends only to those domestic employees with the requisite close personal connection. The application of the household modifier to “companions,” on the other hand, would not effectuate such a distinction. More specifically, given that all companions—even those employed by third-party companies, like Maxim—have the requisite close personal connection, application of the household modifier to the term “companions” would actually have the effect of excluding companions who satisfy the close-personal-connection requirement from the coverage of the companionship exemption. That leads us to the conclusion that we will best give effect to the Division’s intent by construing the household modifier as applying to “domestic employees” but not “companions.”

Our conclusion gains further strength when we consider the particular meaning of the relevant terms. Terms can acquire a particular meaning “by legislative definition or otherwise.” C OLO . R EV . S TAT . § 2-4-101 (emphasis added). For instance, a term may take on a particular meaning in a certain area of law or industry. See, e.g., Wash. Cty. Bd. of Equalization v. Petron Dev. Co., 109 P.3d 146, 153 (Colo. 2005) (concluding that “wellhead” had a particular meaning

29

in the oil-and-gas industry); Armentrout v. FMC Corp., 842 P.2d 175, 186 (Colo. 1992) (noting that “defective” had a particular meaning in design-defect cases). To ascertain the acquired meaning of a term, Colorado courts “may [] look outside the [regulation] to related sources.” People v. Hunter, 307 P.3d 1083, 1086 n.2 (Colo. 2013) (quoting Sullivan v. Indus. Claim Appeals Office, 22 P.3d 535, 538 (Colo. App. 2000)). These related sources include industry guides, informal state agency materials, and statutes and regulations from other jurisdictions. 10 As Maxim points out, the terms “companions,” “casual babysitters,” and “domestic employees” have particular meanings in wage-and-hour law. 11 Both the

10 See, e.g, Hunter, 307 P.3d at 1086 & n.2 (considering “the Sex Offense Management Board’s” handbook persuasive in identifying the intended meaning of “stranger”); Flood v. Mercantile Adjustment Bureau, LLC, 176 P.3d 769, 772 (Colo. 2008) (looking to federal law “for persuasive guidance” on the intended meaning of undefined terms); Askew v. Indus. Claim Appeals Office, 927 P.2d 1333, 1337 (Colo. 1996) (consulting American Medical Association guides for meaning of undefined terms).

11 We note that the Colorado Court of Appeals determined that the portion of the Wage Order exempting “interstate drivers” from the overtime-pay requirement did “not resemble, much less closely parallel, the [Motor Carrier Act] overtime pay exemption to the FLSA.” Brunson, 433 P.3d at 98. Notably, the court observed that whereas “Colorado’s Wage Order lists interstate drivers as exempt employees, the [Motor Carrier Act] overtime pay exemption of the FLSA does not list ‘interstate drivers’ at all.” Id. Given this and other dissimilarities, Brunson thought federal law inapposite when determining the meaning of “interstate drivers” as used in the Wage Order. Instead, the court relied on the definition of “interstate drivers” in the Division’s Bulletin. See id. at 99–100. Here, by contrast, the FLSA itself uses the terms “companionship services” and (continued...)

30

federal regulation operative during times material here and the Division’s Bulletin define “[c]ompanionship services” as those that “provide fellowship, care and protection for a person, who due to advanced age or physical or mental conditions cannot care for his or her own needs.” Aplt.’s App. at 277; see 40 Fed. Reg. at 7405 (language formerly codified at 29 C.F.R. pt. 552, subpt. A, § 552.6). Each source also defines “[c]asual babysitt[ers]” as those who babysit on an “irregular or intermittent” basis and are not full-time or professional babysitters. 29 C.F.R. § 552.5; see Aplt.’s App. at 278. The federal regulation defines “domestic service employment” as “services of a household nature . . . in or about a private home . . . of the person by whom he or she is employed.” 40 Fed. Reg. at 7405 (language formerly codified at 29 C.F.R. pt. 552, subpt. A, § 552.3). The Bulletin, however, does not define “domestic employees.”

Plugging these particular meanings into the companionship exemption indicates that the Division intended to exempt all companions. As in the ordinary sense, in the wage-and-hour sense, companions have a close personal connection with the persons receiving their services. So do casual babysitters. Unlike professional babysitters, casual babysitters are often friends, neighbors, or

11 (...continued)

“babysitting services” “on a casual basis.” See 29 U.S.C. § 213(a)(15). Likewise, as explained below, the Bulletin’s definitions of those terms match the definitions in the operative federal regulation. Thus, unlike in Brunson, we may properly consult federal law to help identify the particular meanings of the relevant terms.

31

relatives of the family. Domestic employees also have a close personal connection with the individuals they serve because, by definition, they “must” work “in or about the private home of the employer.” Id. at 7406 (language formerly codified at 29 C.F.R. pt. 552, subpt. B, § 552.101). In other words, in the federal wage-and-hour context, a domestic employee must work in or about the personal or intimate spaces of the home of the person who hired them—which ordinarily would have obliged that person to make an antecedent decision to personally vest trust in the employee, giving rise to a close personal connection between the employer and the domestic employee. Thus, reasonably assuming that the Division drafted the companionship exemption with these acquired meanings in mind, we may, again, infer that the Division likely intended the companionship exemption to cover those home workers who have a close personal connection with the recipients of their services.

By contrast, we would subvert that intent by reading the household modifier as applying to “companions.” Under that view, the companionship exemption would be underinclusive—not reaching all companions that we may reasonably infer that the Division intended to reach. Specifically, it would not reach otherwise-eligible companions who have the requisite close personal connection with the individuals whom they serve simply because those companions are employed by third-party employers. And, evincing this subversion of the

32

Division’s inferred intent, there would be significant space created by this reading of the scope of the household modifier between the acquired meaning of the term “companion” in the wage-and-hour context—which applies irrespective of the identity of the employer—and the meaning purportedly adopted by the Wage Order’s companionship exemption.

Therefore, the term “companion” of the companionship exemption is best read as not including the household modifier. Put another way, in effectuating the Division’s intent, the term “companion” is best read as including all companions who categorically have the requisite close personal connection with the people whom they serve—even companions employed by third-party employers.

***

In sum, the companionship exemption is ambiguous. One plausible reading is that the exemption applies to all companions. Or it could be read as applying to only those companions “employed by households or family members to perform duties in private residences.” Nonetheless, we conclude that the ordinary and particular meanings of “companions,” “casual babysitters,” and “domestic employees” suggest that the Division intended the former reading. That suggestion, however, is not so strong as to make the Division’s intent “clear.” Specialty, 231 P.3d at 397. Consequently, we may “look to other aids in construction” to pin down the proper scope of the companionship exemption. Id.

33

C

Those other aids in construction confirm what the ordinary and particular meanings suggest: the companionship exemption applies to all companions. Two interpretive aids—the presumption that agency actions are valid, and the surplusage canon—are instructive. Each counsels in favor of reading the companionship exemption as applying to all companions. Although it is true, as the district court emphasized, that the series-qualifier canon points in the other direction, that canon cannot bear the interpretive weight that the district court put on it. As a result, we agree with Maxim that the companionship exemption applies to all companions—in particular, a third-party employer such as Maxim.

1

In Colorado, “agencies are legally bound to comply strictly with their enabling statutes.” Schlapp ex rel. Schlapp v. Colo. Dep’t of Health Care Policy & Fin., 284 P.3d 177, 182 (Colo. App. 2012) (quoting Adams v. Colo. Dep’t of Social Servs., 824 P.2d 83, 86 (Colo. App. 1991)); accord Maul v. State Bd. of Dental Exam’rs, 668 P.2d 933, 937 (Colo. 1983). A regulation that goes beyond the authority conferred by its enabling statute is void. See C OLO . R EV . S TAT . § 24-4-106(7)(b) (“The court shall hold unlawful and set aside the agency action” if it exceeds “statutory . . . authority . . . .”). But Colorado courts “assume an agency action is valid.” Schlapp, 284 P.3d at 180; see State Bd. of Chiropractic

34

Exam’rs v. Stjernholm, 935 P.2d 959, 972 (Colo. 1997) (“Actions of state agencies have a presumption of validity and regularity.”). Consistent with that assumption, courts prefer to interpret regulations to avoid conflicts with enabling statutes. See Schlapp, 284 P.3d at 180 (“[W]e interpret a regulation so as not to conflict with the objective of the statute it implements.”); Bd. of Cty. Comm’rs v. BDS Int’l, LLC, 159 P.3d 773, 779 (Colo. App. 2006) (“[W]e will construe the County Regulations, if possible, so as to harmonize them with the applicable state statutes . . . .”); see also Hanien v. Gessler, 333 P.3d 41, 54 (Colo. 2014) (Eid, J., dissenting) (“[W]e are obligated to interpret regulations and statutory provisions as a consistent, harmonious whole where possible, not to maximize discord . . . .”). In other words, Colorado courts generally follow the widely held maxim that “[a]n interpretation that validates outweighs one that invalidates.” S CALIA & G ARNER , supra, at 66; see Lobato v. Indus. Claim Appeals Office, 105 P.3d 220, 224 (Colo. 2005) (“If statutory provisions are in conflict, we will adopt the interpretation that best harmonizes the provisions if possible.”).

The enabling statute here authorizes the Division to issue wage orders setting the “conditions and rules” for overtime pay. C OLO . R EV . S TAT . § 8-6- 111(4). But the statute specifies that those conditions and rules “shall apply equally to all employers in [a given] industry or occupation.” Id. Hence, a wage order requiring Restaurant A to pay its wait staff overtime wages but exempting

35

the wait staff employed by Restaurant B from that requirement ordinarily would violate the enabling statute because it would treat employers in the same industry (i.e., restaurants) differently.

In a similar way, Ms. Jordan’s (and the district court’s) interpretation of the companionship exemption violates the enabling statute. Under her reading of it, the companionship exemption requires third-party employers to pay their companions overtime wages, while it simultaneously exempts “households or family members” who employ companions to perform like duties. In short, her interpretation runs afoul of the enabling statute because it treats employers in the same occupation or industry (i.e, companionship services) unequally. In other words, by virtue of its unequal application to employers of companions, Ms. Jordan’s interpretation essentially renders the term “companion” void by operation of the enabling statute. And Ms. Jordan does not even try to reconcile this conflict with the enabling statute. Thus, her reading of the companionship exemption is one to eschew. Under Colorado law, we should instead adopt a reasonable interpretation that better harmonizes the companionship exemption with the enabling statute, and Maxim provides such a reasonable interpretation—specifically, one that interprets the companionship exemption as applying to all companions, irrespective of the identity of their employer.

36

To be sure, even under Maxim’s construction of the companionship exemption, there is space for some conflict with the enabling statute. As noted supra, that statute commands that overtime rules “shall apply equally to all employers in [a given] industry or occupation.” Id. And the companionship exemption exempts three types of employees—“companions,” “casual babysitters,” and “domestic employees”—from the overtime-pay requirement. Were we to adopt Maxim’s reading, the companionship exemption only would apply evenly to employers of two of the three employee types—“companions” and “casual babysitters”—but not to the third type, “domestic employees.” More specifically, under Maxim’s reading of the companionship exemption, the household modifier would apply solely to the employee category of “domestic employees,” thereby allowing for an unequal application of the companionship exemption to employers of domestic employees. That is, third-party employers of domestic employees would not be entitled to invoke the companionship exemption because of the effect of the household modifier; rather, only “households or family members” who employ domestic employees “to perform duties in private residences” would be able to do so.

But this unequal application of the companionship exemption to employers of domestic employees—which creates a basis for conflict with the enabling statute—is reasonably viewed as simply a function of the Division’s inclusion of

37

the household modifier in the companionship exemption in the first place, rather than as a result of any misstep on Maxim’s end in interpreting the companionship exemption. Stated differently, the household modifier is, by its plain terms, restrictive; its language is limited to one subset of employers, and one subset of employers only: “households or family members” who employ workers “to perform duties in private residences.” And, given the Division’s inclusion of the household modifier in the companionship exemption—if courts are to remain faithful to the interpretive principles of Colorado law, which require, to the extent possible, that all the terms of the agency’s regulations be given effect (as we discuss further infra)—then the operative question is not whether the household- modifier’s restriction applies at all to the employers that the companionship exemption implicates but rather to which of those employers does the modifier apply. And that question necessarily requires that we consider, as we do here, which of those classes of employers may be subjected to unequal treatment by the application of the household modifier to them, such that this application may engender a conflict with the enabling statute.

And even if Maxim’s interpretation of the companionship exemption does allow for some conflict with the enabling statute by restricting the application of the household modifier solely to domestic employees, its interpretation would nevertheless allow for less of a restriction—and thus present less of a

38

conflict—than would Ms. Jordan’s reading. Under Ms. Jordan’s reading of the companionship exemption, the household modifier would apply to employers of all three types of employees that the exemption covers. And because this could at least arguably result in unequal treatment of employers of all three employee types—viz., treatment that excludes third-party employers of all three from being able to invoke the exemption—the extent of the exemption’s possible conflict with the enabling statute would be greater under Ms. Jordan’s reading. The upshot of this is that under Ms. Jordan’s reading, all three employee classifications could possibly be rendered void because her reading allows for unequal treatment of their respective employers. Under Maxim’s interpretation, by contrast, only one employee classification, domestic employees, could possibly suffer this fate. 12 And that one void portion of the regulation would “not affect

12 We note that the federal definition of “casual babysitters” has built into it a limited universe of employers, and this definition conceivably could have implications for the proper resolution of the kind of interpretive issues addressed by this opinion, including the unequal-application issue under the enabling statute, which we discuss supra. Specifically, under federal law, babysitters “who are employed by an employer or agency other than the family or household using their services are not employed on a ‘casual basis.’” 29 C.F.R. § 552.109(b) (emphasis added). In other words, by definition, casual babysitters under federal law are employed by families or households—not by third-party employers. If the Wage Order mirrors federal law in this respect (a matter that we do not decide), then this limitation based on employer type—which, again, is built into the definition of casual babysitter—could have some bearing, even if only to a minor extent, on the resolution of the interpretive issues before us. One such possible effect of the companionship exemption’s coverage of only casual babysitters (as opposed to babysitters more generally) could be that only a certain subset of (continued...)

39

the validity of the remaining portion of the Wage Order.” 7 C OLO . C ODE R EGS . § 1103-1:14; cf. People v. Tate, 352 P.3d 959, 975 (Colo. 2015) (Rice, C.J., concurring in part and dissenting in part) (“[S]evering portions of a statute is preferred over invalidating the entire statute . . . .”).

To be clear, our purpose here is not to opine on whether any portion of the companionship exemption should be deemed void under Colorado law, and we offer no opinion on that matter. 13 Rather, our aim is to determine which interpretation of the ambiguous companionship exemption under the circumstance of this case is the best one. And, as between Maxim’s interpretation, which could possibly result in one employee classification of the exemption being declared void, and Ms. Jordan’s, which could possibly result in all of the employee

12 (...continued)

employers would be able to avail itself of this exemption: households or families would be free to do so, while third-party employers would be categorically precluded from invoking the companionship exemption applicable to casual babysitters. Does such an employer-type limitation raise unequal-application concerns under the enabling statute? Well, neither the parties nor the district court has expressly addressed the possible implications (if any) that this employer-type limitation may have on the unequal-application issue or on any other interpretive issue addressed by this opinion, for that matter. Therefore, we decline to address these possible implications here, let alone definitively opine on them.

13 Ms. Jordan does not argue that Maxim’s reading violates the enabling statute, nor did the district court conclude (or suggest) as much. Maxim, however, does allege that Ms. Jordan’s reading violates the enabling statute. And Ms. Jordan greets that allegation with silence.

40

classifications being declared void, we believe Maxim’s interpretation is the better one.

2

Another interpretive tool—the surplusage canon—makes our decision to adopt Maxim’s interpretation that much easier. In Colorado, courts “disfavor constructions that render [regulatory] language superfluous.” Kinder Morgan CO 2 Co. v. Montezuma Cty. Bd. of Comm’rs, 396 P.3d 657, 667 (Colo. 2017). Applying this canon, the district court concluded that Maxim’s reading renders the “and” in “companions, casual babysitters, and domestic employees” superfluous. On appeal, Maxim turns the surplusage canon against the district court and argues that its interpretation makes “companions” and “casual babysitters” redundant. We think Maxim’s argument carries the day. 14

14 For her part, Ms. Jordan never once mentions the surplusage canon on appeal. The closest she comes to doing so is in perfunctorily noting that “the District Court correctly rejected [Maxim’s] flawed reading” because that reading “would have required the [c]ourt to ignore part of the language contained in the provision.” Aplee.’s Resp. Br. at 10. This oblique nod to the surplusage canon is likely insufficient for preservation purposes. See Marshall v. Columbia Lea Reg’l Hosp., 474 F.3d 733, 739 (10th Cir. 2007) (declining to consider an argument not raised in appellee’s response brief). That said, “we may affirm on any basis supported by the record.” Richison v. Ernest Grp., Inc., 634 F.3d 1123, 1130 (10th Cir. 2011). “In exercising that discretion [the court] consider[s] whether the ground was fully briefed and argued here and below, whether the parties have had a fair opportunity to develop the factual record, and whether, in light of factual findings to which we defer or uncontested facts, [the court’s] decision would involve only questions of law.” Feinberg v. Comm’r, 916 F.3d 1330, 1334 (10th Cir. 2019) (alterations in original) (quoting Elkins v. Comfort, 392 F.3d (continued...)

41

We start with the allegedly superfluous “and” that troubled the district court. That court determined that if the Division wanted the household modifier to apply only to “domestic employees,” the Division would not have inserted an “and” before “domestic employees.” That is to say, if the Division had intended Maxim’s reading to apply, it would have written the following: “[o]ther exemptions include: companions, casual babysitters . . . , domestic employees employed by households or family members to perform duties in private residences, property managers . . . , and bona fide volunteers.” But the Division did not write the companionship exemption that way. So, to give the “and” before “domestic employees” meaning, under the district court’s reasoning, the household modifier should be read as applying to “companions” and “casual babysitters,” as well as to “domestic employees.”

However, even if Maxim’s interpretation does create an unnecessary “and,”

it gives us no cause for concern. 15 Above all else, our job “is to ‘give effect to

14 (...continued)

1159, 1162 (10th Cir. 2004)). Applying that discretion here, in particular, we find that the latter two Feinberg factors strongly militate in favor of our careful examination of whether the district court’s rationale based on the surplusage canon is sound and supportive of its judgment: the parties have had a fair opportunity to develop the factual record, and our analysis turns on the purely legal question of whether or not the surplusage canon provides cogent aid for interpreting the companionship exemption as applying to third-party employers of companions.

15 Because Maxim does not argue otherwise, we assume without (continued...)

42

the intent of the enacting body.’” Richter, 796 F.3d at 1185 (quoting Benuishis, 195 P.3d at 1145). The Division does not appear to have been “concerned about redundant conjunctive language.” Aplt.’s Opening Br. at 31. Had it been, it would have drafted section 5 as “a single long list of exempt employees and occupations, with a single ‘and’ at the end.” Id. But it did not do so. Instead, the Division used seven “ands”—at least four of which would be unnecessary in one long list. See, e.g., 7 C OLO . C ODE R EGS . § 1103-1:5 (“Also exempt are: students employed by sororities, fraternities, college clubs, or dormitories, and students employed in a work experience study program and employees working in laundries . . . .” (emphases added)). The Division also littered section 5 with unnecessary connective phrases. That section begins, “[t]he following employees or occupations . . . are exempt.” Id. It then adds the connective phrases, “[o]ther exemptions are” and “[a]lso exempt are.” These phrases strongly bespeak redundancy. Thus, the Division clearly intended, for clarity-of-exposition purposes or otherwise, a certain amount of “redundant conjunctive language.” Aplt.’s Opening Br. at 31. Given that intent, a redundant “and” does not undercut the persuasiveness of Maxim’s reading of the companionship exemption. See Benefield v. Colo. Republican Party, 329 P.3d 262, 267 (Colo. 2014) (“[T]he

15 (...continued)

deciding that its reading in fact renders the “and” before “domestic employees” superfluous.

43

superfluity principle, like all canons of construction, is merely an interpretive aid, not an absolute rule.”); S CALIA & G ARNER , supra, at 176 (“[The surplusage canon] cannot always be dispositive because . . . the underlying proposition is not invariably true. Sometimes drafters do repeat themselves and do include words that add nothing of substance . . . .”).

By contrast, surplusage stemming from Ms. Jordan’s reading is cause for concern. Recall that the ordinary meaning of “domestic employee” is a person employed to perform services “relating to the family or household.” Domestic, A MERICAN H ERITAGE , supra, at 550. Companions and casual babysitters fit within that definition: they perform services relating to the family or household; that is, those two terms are subsets of domestic employees. Accordingly, if the household modifier were to apply not only to “domestic employees,” but also to “companions” and “casual babysitters,” the latter two terms would be rendered redundant. Specifically, the phrase “domestic employees employed by households or family members to perform duties in private residences” would subsume “companions,” in that companions also would be domestic employees employed only by “households or family members” to perform such duties. And in a similar vein, that phrase would likewise subsume “casual babysitters,” who also would be domestic employees employed only by “households or family members” to carry out such duties. To give “companions” and “casual babysitters” any possibility of

44

having effect, then, we should read the household modifier as only applying to “domestic employees.” In particular, and of central importance here, this reading unquestionably would permit the term “companion” to have effect. Cf. Lockhart v. United States, 136 S. Ct. 958, 966 (2016) (preferring the interpretation creating less surplusage); United States v. Atl. Research Corp., 551 U.S. 128, 137 (2007) (“It is appropriate to tolerate a degree of surplusage rather than adopt a textually dubious construction that threatens to render the entire provision a nullity.”); Kinder Morgan, 396 P.3d at 667 (rejecting interpretation rendering an entire statutory clause superfluous); Benefield, 329 P.3d at 265 (preferring reading that arguably rendered one word redundant over a reading that created more substantial surplusage). In other words, if the household modifier is limited to “domestic employees,” then the term “companions”—although still involving the performance of services relating to the family or household, like other domestic employees—could have a distinct meaning: it would contemplate employers other than “households or family members” (i.e., employers other than those addressed by the household modifier)—specifically, third-party employers such as Maxim.

Accordingly, the surplusage canon favors Maxim’s reading. Even if that reading renders the “and” before “domestic employees” superfluous, we much prefer an extra “and” to the meaningless presence in the companionship exemption of the terms “companions” and “casual babysitters”—especially when

45

the former of those two terms is the exact one at issue here. Thus, as with our enabling-statute analysis, we conclude that the surplusage canon militates in favor of interpreting the companionship exemption as applying to all companions—including those employed by third-party employers, like Maxim.

3

A centerpiece of the district court’s contrary interpretation of the companionship exemption, however, was the series-qualifier canon. That canon teaches that “[w]hen a referential or qualifying clause follows several words or phases and is applicable as much to the first word or phrase as to the others in the list, . . . the clause should be applied to all of the words or phrases that preceded it.” Aplt.’s App. at 289 (omission in original) (quoting Kennett, 135 F. Supp. 3d at 1241). Without explanation, the district court thought “the household qualifier [] equally applicable to the antecedents ‘companions’ and ‘casual babysitters’ as it [was] to ‘domestic employees.’” Id. (quoting Kennett, 135 F. Supp. 3d at 1241). So it followed the series-qualifier canon and concluded that the companionship exemption did “not extend to third-party employers.” Id. at 290.

In our view, the district court should have declined to apply the series-

qualifier canon here. “Perhaps more than most of the other canons, this one is highly sensitive to context.” S CALIA & G ARNER , supra, at 150. “Often the sense of the matter prevails” over the meaning the series-qualifier canon suggests. Id.

46

“He went forth and wept bitterly,” for example, “does not suggest that he went forth bitterly.” Id. Similarly, the series-qualifier canon, “perhaps more than most [canons] . . . [,] is subject to defeasance by other canons”—that is, it is perhaps more prone than most to have its effect nullified by other canons. Id. At bottom, “like all canons of construction, [the series-qualifier canon] is merely an interpretive aid, not an absolute rule.” Benefield, 329 P.3d at 267. Its utility depends on “context and consideration of other, and often conflicting, interpretive aids.” Id.

In our view, the context is not suitable for application of the series-

qualifier canon and, accordingly, the district court should have declined to apply it. Remember, the ordinary and particular meanings of the terms suggest that the Division intended the companionship exemption to apply to those household workers who have a close personal connection to the recipients of their services. Application of the series-qualifier canon would undermine that intent by rendering the companionship exemption underinclusive: it would not cover companions employed by third-party employers—even though those companions have the requisite close personal connection with those they serve and, consequently, are the type of household workers that the Division intended to cover under the companionship exemption. The aptness of the series-qualifier canon in this context is further undermined by the other canons of construction.

47

Most notably, using the series-qualifier canon here creates conflict with the enabling statute, which expressly requires that any overtime requirements for particular categories of employees “apply equally to all employers in such industry or occupation.” C OLO . R EV . S TAT . § 8-6-111(4). In particular, as most relevant here, applying the household modifier to the term “companions”—in line with the series-qualifier canon—would allow for the differential application of the companionship exemption to employers in the same industry or occupation, contrary to the mandate of the enabling statute. Not only that, but application of the series-qualifier canon also gives rise to surplusage. Given these signs that the context is not suitable for application of the series-qualifier canon, we believe the district court should have declined to apply it.

Lockhart v. United States nicely illustrates the perils of blind allegiance to the series-qualifier canon. 16 At issue there was a section of the Federal Criminal

16 The Supreme Court decided Lockhart after Maxim had filed its reply in support of its motion to dismiss. In its subsequently filed cross-motion for summary judgment, Maxim cited Lockhart as support for its interpretive argument. The district court, however, thought it “improper” to consider new interpretive arguments not raised at the motion-to-dismiss stage. Aplt.’s App. at 585 (Order Granting in Part & Den. in Part Pl.’s Mot. for Partial Summ. J. & Def.’s Cross Mot. for Summ. J., dated Mar. 29, 2018). On appeal, Maxim reprises its Lockhart argument. Ms. Jordan does not engage with this argument or assert that Maxim has failed to preserve it. See generally Aplee.’s Resp. Br. (failing to cite Lockhart or raise preservation concerns). Given Ms. Jordan’s silence, even assuming Maxim failed to preserve its Lockhart argument before the district court (a question we do not reach), we overlook this possible failing. See Cook v. Rockwell Int’l Corp., 618 F.3d 1127, 1138–39 (10th Cir. 2010)

(continued...)

48

Code that increased the prison sentence for defendants convicted of possessing child pornography “if they ha[d] ‘a prior conviction . . . under the laws of any State relating to aggravated sexual abuse, sexual abuse, or abusive sexual conduct involving a minor or ward.” 136 S. Ct. at 961 (omission in original) (quoting 18 U.S.C. § 2252(b)(2)). The question was “whether the phrase ‘involving a minor or ward’ modifie[d] all items in the list of predicate crimes . . . or only the one item that immediately precede[d] it.” Id. The Court thought the latter, holding that “‘involving a minor or ward’ modifie[d] only ‘abusive sexual conduct,’ the antecedent immediately preceding it.” Id. at 962.

In reaching that holding, the Court offered three reasons that the series-

qualifier canon could not “bear the weight that either Lockhart or the dissent” attributed to it. Id. at 965. First, applying that canon “would risk running headlong into the rule against superfluity.” Id. at 965–66. If the phrase “involving a minor or ward” were to modify not just “abusive sexual conduct,” but “aggravated sexual abuse” and “sexual abuse” as well, the Court explained, those latter two terms would become “hopelessly redundant.” Id. at 965. Applying the modifier to only “abusive sexual conduct,” by contrast, “preserve[d] some distinction between the categories.” Id. at 966. The Court recognized that

16 (...continued)

(considering argument forfeited by a party because the other party “forfeited any forfeiture argument [it] may have [had]”).

49

its preferred reading did “not eliminate all superfluity,” but the series-qualifier canon created even more surplusage, which counseled against applying it. See id.

Second, the Court declined to follow the series-qualifier canon because of the strong “parallel” between three other sections of the Federal Criminal Code—which applied the phrase “involving a minor or ward” to only the antecedent “abusive sexual conduct”—and the instant section of the Federal Criminal Code. In particular, the Court noted that those three other sections—entitled, respectively, “Aggravated sexual abuse,” “Sexual abuse,” and “Sexual abuse of a minor or ward”—“mirror[ed] precisely” the “structure and language” of the section’s text at issue here: “aggravated sexual abuse, sexual abuse, or abusive sexual conduct involving a minor or ward.” Id. at 964. With this fact in mind, the Court “doubt[ed]” that if “Congress had intended to limit” the terms “aggravated sexual abuse” and “sexual abuse” “to conduct ‘involving a minor or ward,’” that it would have “so closely [followed] the structure and language” of those other sections of the Code. Id.

Third, the Supreme Court considered the last-antecedent canon a better fit.

That canon “provides that ‘a limiting clause or phrase . . . should ordinarily be read as modifying only the noun or phrase that it immediately follows.’” Id. at 962 (alteration in original) (quoting Barnhart v. Thomas, 540 U.S. 20, 26 (2003)). According to the Court, following the last-antecedent canon made more sense

50

because the provision did “not contain items that readers are used to seeing listed together or a concluding modifier that readers are accustomed to applying to each of them.” Id. at 963. But at bottom, the Court reminded the parties that whether a modifier applies to each word in a list or only to the last word is a “fundamentally contextual question[].” Id. at 965. And the last-antecedent canon was just another piece of the context that convinced the Court that the modifier applied only to the last word in the list.

Lockhart’s first two lessons map onto this case. As in Lockhart, following the series-qualifier canon here creates (for reasons explicated supra) serious surplusage; it makes “companions” and “casual babysitters” redundant with “domestic employees.” Likewise, the Division seems to have borrowed the terms “companions” and “casual babysitters” from federal wage-and-hour law. And under that law, it is clear that companions need not have been employed by a family or household in order to be exempt from overtime requirements—viz., they could be employed by third-party employers. See 40 Fed. Reg. at 7407 (language formerly codified at 29 C.F.R. pt. 552, subpt. B, § 552.109(a)). As in Lockhart, then, it would be odd for the Division to have used the term “companion” in the companionship exemption, if it had intended the term to not contemplate third- party employers, when the likely touchstone of federal wage-and-hour law

51

contemplated such employers. Thus, Lockhart’s first two lessons counsel against applying the series-qualifier canon here.

That said, Lockhart’s third lesson—the last-antecedent canon—is one we cannot embrace. The Colorado General Assembly has “declare[d]” that the last- antecedent canon “has not been adopted by the general assembly and does not create any presumption of statutory intent.” C OLO . R EV . S TAT . § 2-4-214. We therefore cannot add that canon to the already lengthy list of reasons to part from the series-qualifier canon.

Maxim begs to differ. It argues that the anti-last-antecedent statute does not apply here because this case involves “interpret[ing] a regulation adopted by the Division of Labor, not a statute adopted by the General Assembly.” Aplt.’s Opening Br. at 36. For support, Maxim cites a Colorado case applying the forbidden last-antecedent canon to interpret a contract. See id. (citing Chandler- McPhail v. Duffey, 194 P.3d 434, 440–41 (Colo. App. 2008)).

But beyond splitting hairs, Maxim’s argument is misguided. For starters, Maxim concedes that the same “basic rules of interpretation which pertain to the construction of a statute” also apply to regulations like the Wage Order. Aplt.’s Opening Br. at 26 (quoting Richter, 796 F.3d at 1185). Thus, if the General Assembly has removed the last-antecedent rule from the statutory-interpretation toolbox, as it has, that rule can no longer be used to interpret regulations either.

52

And Maxim’s invocation of Duffey does not alter the analysis. The court there overlooked the anti-last-antecedent statute because the meaning of a contract between private parties turns on “the intent of the contracting parties.” 194 P.3d at 441. By contrast, the meaning of a regulation turns on the promulgating agency’s intent, which is, in effect, an extension of the General Assembly’s will. That is why Colorado courts must “set aside [an] agency action” if it exceeds “statutory authority.” C OLO . R EV . S TAT . § 24-4-106(7)(b). And so ignoring the anti-last-antecedent statute made sense in Duffey because the General Assembly’s intent was not implicated. The same is not true here. Thus, we will adhere to the General Assembly’s declaration that the last-antecedent canon is inoperative.

Even without the last-antecedent canon at our disposal, for the reasons already discussed, the series-qualifier canon is not a good fit in this context. Furthermore, People v. Weeks, 369 P.3d 699 (Colo. App. 2015), a case that Maxim cites, does not lead us to a different view. The statute there defined “child abuse” as “a continued pattern of conduct that results in malnourishment, lack of proper medical care, cruel punishment, mistreatment, or an accumulation of injuries that ultimately results in the death of a child or serious bodily injury to a child.” Id. at 710–11 (quoting C OLO . R EV . S TAT . § 18-6-401(1)(a)). The question was whether the phrase “ultimately results in the death of a child or serious bodily injury” applied to “each of the enumerated ‘patterns of conduct’”

53

or “only to the last pattern of conduct.” Id. at 711. The Weeks court noted that it could not rely on the last-antecedent canon. But it also explained that it need not follow the series-qualifier canon and “presume that the last phrase applie[d] to all of the previously enumerated patterns of abuse.” Id. Accordingly, the court answered the interpretive question by consulting the statute’s context. And that context led the court to hold that “the last phrase ‘ultimately results in the death of a child or serious bodily injury to a child’ applie[d] to only the last enumerated pattern of abuse.” Id. at 712. In sum, notwithstanding the anti-last-antecedent statute, Weeks declined to follow the series-qualifier canon.

People v. Trujillo, 251 P.3d 477 (Colo. App. 2010) did the same. The statute there rendered certain statements made “by a child” admissible in “proceedings in which a child is a victim of an unlawful sexual offense . . . or is a victim of incest . . . when the victim was less than fifteen years of age at the time of the commission of the offense.” Id. at 479 (quoting C OLO . R EV . S TAT . § 13-25- 129 (2009)). Invoking the series-qualifier canon, the defendant argued that the age modifier must apply “both to victims of incest and to victims of an unlawful sexual offense.” Id. In a bid to bolster this argument, the defendant relied on the anti-last-antecedent statute. But that reliance was “misplaced,” the court explained, because that statute did not compel Colorado courts to follow the series-qualifier canon. Id. at 480. Rather, when a modifier could apply either to

54

only the last item in a list or to each listed, courts should still “turn to principles of statutory interpretation.” Id. One such principle, the court emphasized, was the need to “give effect to every provision of a statute while avoiding conflicts.” Id. at 480–81. Applying the fifteen-year-old-age modifier to “an unlawful sexual offense” would violate both parts of that maxim. Id. First, it would conflict with the provision defining “unlawful sexual offense” as including “offenses involving child victims older than fifteen years of age.” Id. at 481. Second, it would render surplusage the statute’s proviso that “the statutes which are the subject of the action” provide the definition of “child.” Id. Unwilling to elevate the series- qualifier canon above all others, the court held that the phrase “when the victim was less than fifteen years of age” applied only to “a victim of incest” and not to “a victim of an unlawful sexual offense.” Id.

Trujillo and Weeks teach that Colorado courts are free—and, thus, so are we—to eschew the series-qualifier canon and instead follow the context and other canons. As explained above, the context and other canons militate in favor of reading the household modifier as applying only to “domestic employees.” We therefore decline to apply the series-qualifier canon to the companionship exemption.

A comparison between our case and the two cases the district court cited in support of applying the series-qualifier canon reinforces our decision to eschew

55

that canon. Take Estate of David v. Snelson, 776 P.2d 813 (Colo. 1989). At issue in that case was whether two adopted children could inherit from their natural father’s estate. The governing statute there provided that adopted children could inherit from their natural parents “except to the extent that inheritance rights have been divested by a final order of relinquishment, a final decree of adoption, or an order terminating the parent-child relationship under the laws of this state or of any other jurisdiction.” Id. at 816 (quoting C OLO . R EV . S TAT . § 15-11-109(1)(a) (1987)). Consistent with that statute, the Snelson court explained that whether the sons could inherit from their natural father’s estate “depend[ed] on the meaning of the statutory phrase ‘except to the extent that inheritance rights have been divested by . . . a final decree of adoption . . . under the laws of this state or of any other jurisdiction.” Id. In undertaking this interpretive inquiry, the court used various canons of construction. It began with the series-qualifier canon, observing that the language “under the laws of this state or of any other jurisdiction” was “as applicable to the first phrases as to the others in the list,” and as such, that language modified not only “an order terminating the parent- child relationship,” but also “a final order of relinquishment” and, as relevant there, “a final decree of adoption.” Id. at 818. The court’s analysis did not stop with the series-qualifier canon, however. It next turned to the surplusage canon, and it was this canon to which it devoted the most airtime. The court reasoned

56

that construing the statute to require an examination of only the adoption decree’s language itself, as the sons urged, would, among other issues, give no effect to the phrase “under the laws of this state or of any other jurisdiction,” thereby flouting the surplusage canon’s command that “a statute must be construed whenever possible to give effect to all of its parts.” Id. at 818–19. The court then went on to examine various other problems that the sons’ proffered reading posed under the surplusage canon, followed by a fulsome explication of why construing the statute to require an examination of the laws in effect at the time of adoption was a sensible reading under the canon. See id. at 819.

The long and short of it is this: while the series-qualifier canon did figure into the parties’ arguments and the Snelson court’s analysis, it was hardly pivotal. Indeed, we note that there is no indication in Snelson’s text that the parties “focused on” the modifier “under the laws of this state or any other jurisdiction” and argued against the application of the series-qualifier canon. See id. at 816 n.2. Here, by contrast, Ms. Jordan hangs her hat on this canon (despite offering no defense of why its application would be proper here), while Maxim vigorously argues against it. And yet another point of distinction is that in Snelson, the court’s discussion centered in substantial part around the surplusage canon. And under that canon, it concluded that construing the modifying phrase at issue to apply to all of the phrases in the series, as opposed to only the phrase immediately

57

preceding that modifier, would avoid considerable surplusage. See id. at 818–19. Doing so here, on the other hand, for reasons explicated supra, would create surplusage. Accordingly, unlike the district court, we do not view the circumstances here as providing an appropriate context for the application of the series-qualifier canon.

People ex rel. M.W., 796 P.2d 66 (Colo. App. 1990) likewise fails to convince us to follow the series-qualifier canon. The statute there required a guardian ad litem to “examine and cross-examine witnesses in both adjudicatory and dispositional hearings, introduce and examine his own witnesses, make recommendations to the court concerning the child’s welfare, appeal matters to the court of appeals . . . , and participate further in the proceedings to the degree necessary to adequately represent the child.” Id. at 68 (emphasis added) (quoting C OLO . R EV . S TAT . § 19-3-203(3) (1989)). A mother alleged that the trial court had wrongly terminated her parent-child relationship because, although “the guardian ad litem took an active part in the proceedings, he did not make any recommendation” to the court concerning the child’s welfare. Id. at 67. The court of appeals rejected that argument. Applying the series-qualifier canon, the court concluded that the modifier “to the degree necessary to adequately represent the child” extended to the entire list. Id. at 68. Hence, a guardian ad litem need not always “make a recommendation to the trial court in a termination hearing”

58

and the trial court thus did not err in terminating the parent-child relationship without such a recommendation. Id.

Again, the court’s application of the series-qualifier canon in M.W. made sense. To have eschewed the canon would have yielded odd results. Indeed, if the court were to have applied the modifier only to the instruction for guardians ad litem to “participate further in the proceedings,” then a guardian ad litem would always have had to “appeal matters to the court of appeals”—even if the guardian thought doing so was contrary to the child’s best interest. Id. at 68. No such odd results arise here from reading the household modifier to apply only to “domestic employees.” Put simply, the mere fact that M.W. applied the series- qualifier canon there does not justify applying it under the distinct circumstances of this case.

No doubt under a purely mechanical application of the series-qualifier canon, the household modifier would apply to “companions.” But again, this canon “is highly sensitive to context.” S CALIA & G ARNER , supra, at 150. And the context here makes plain that the Division intended the companionship exemption to apply to all companions—including those employed by third-party employers.

***

The foregoing analysis is sufficient to resolve this case. It imparts that the companionship exemption is ambiguous but that the ordinary and particular

59

meanings of “companions,” “casual babysitters,” and “domestic employees,” along with other tools of construction, compel the conclusion that the companionship exemption applies to all companions—irrespective of whether their employers are households or family members on the one hand, or third-party companies, like Maxim, on the other. But we find persuasive support for our conclusion in the Division’s longstanding and consistent interpretation of the companionship exemption. An examination of that interpretation therefore follows.

D

Explaining how and why we find the Division’s interpretation persuasive is a three-step process. In step one, we recount how the Division has interpreted the companionship exemption. In step two, we decline to decide whether any formal deference attaches to the Division’s interpretation; instead, we conclude that we may treat that interpretation as at least persuasive authority. In step three, we explain why we find that interpretation persuasive.

1

The Division has, as is relevant here, interpreted the companionship exemption in two sources: (1) two substantively identical opinion letters to Maxim, and (2) two administrative proceedings. In both sources, the Division

60

confirmed that the companionship exemption applies to companions employed by third-party employers.

We start with the opinion letters to Maxim. In both letters, the Division stated its “position . . . that the treatment and interpretation of the companions exemption in the Wage Order w[ere] intended to mirror the companions definition and associated regulations contained in federal law.” Aplt.’s App. at 189, 192. It added that “this position has remained unchanged since . . . 1998.” Aplt.’s App. at 189, 192. The Division then quoted the then-extant federal regulation: “[e]mployees who are engaged in providing companionship services . . . and who are employed by an employer or agency other than the family or household using their services, are exempt from the [FLSA’s] minimum wage and overtime pay requirements . . . .” Id. at 189, 193 (quoting 40 Fed. Reg. at 7407 (language formerly codified at 29 C.F.R. pt. 552, subpt. B, § 552.109(a))). Leaving nothing to implication, the Division ended the letters by stating its “practice of applying the companionship exemption in situations involving third party employers.” Id. at 190, 193.

The two administrative proceedings, which were instituted in 2012 and 2014, put the interpretation announced in the opinion letters into practice. In the 2012 proceeding, a “companion” claimed that Maxim owed the employee overtime wages. Id. at 539–40. The Division denied that claim. In doing so, it

61

reiterated that Colorado’s companionship exemption was “intended to mimic federal law,” which at the time provided that the exemption applied to companions working for third-party employers. Id. at 541. Thus, the Division concluded that it lacked “the authority to pursue th[e] matter further under Colorado law.” Id. Put differently, the Division determined that the Maxim companion was exempt from overtime wages under Colorado law.

The Division stuck to its longstanding interpretation in the 2014 administrative proceeding. Again, a Maxim employee filed a complaint with the Division. This time, though, the employee explicitly argued that the companionship exemption applied to only those companions “employed directly by households,” not to those employed by third-party employers such as Maxim. Id. at 544. The Division rejected that argument and reaffirmed that “the companions exemption in the Wage Order [was] intended to mirror the current definition and associated regulations in federal law.” Id. at 546. Therefore, once again, the Division dismissed the complaint for lack of “statutory authority to pursue th[e] matter further.” Id.

The bottom line is this: since 1998, the Division has consistently interpreted the companionship exemption as applying to companions employed by third-party employers. The Division has reiterated this longstanding interpretation in at least two opinion letters. And it has put its interpretation to

62

work in at least two administrative proceedings. Simply put, the Division’s words and actions confirm that it intended the companionship exemption to apply to companions employed by third-party employers.

2

On appeal, the parties quarrel over whether we must defer to the Division’s longstanding interpretation. Maxim thinks so. It claims that under Colorado law, the Division’s interpretation “easily merits deference,” as it is neither “‘plainly erroneous [n]or inconsistent with’ the Wage Orders’ text.” Aplt.’s Opening Br. at 43–44 (quoting Schneider v. Indus. Comm’n, 624 P.2d 371, 373 (Colo. App. 1981)). Ms. Jordan disagrees. She argues that deference is inappropriate here because the Division’s interpretation contradicts the companionship exemption’s “plain language.” Aplee.’s Resp. Br. at 17.

We need not decide whether any deference attaches to the Division’s interpretation 17 because under Colorado law, it is undisputed that we may treat the

17 Maxim’s argument that the Division’s interpretation proffered in the administrative proceedings merits deference may have preservation problems. In particular, Maxim did not even mention the administrative proceedings at the motion-to-dismiss stage. Nor did Maxim argue that those proceedings merited deference in its cross-motion for summary judgment. Perhaps that is why the district court never considered deferring to those proceedings in its order granting summary judgment in Ms. Jordan’s favor. That said, because we neither accept nor reject either party’s arguments regarding deference, and given Ms. Jordan’s failure to raise any preservation concerns on appeal, any preservation problems are of no moment.

63

Division’s interpretation of its Wage Order as persuasive authority. 18 After all, Colorado’s cardinal rule of interpretation “is to ‘give effect to the intent of the enacting body.’” Richter, 796 F.3d at 1185 (quoting Benuishis, 195 P.3d at 1145). To help in this endeavor, the General Assembly allows courts to consider “[t]he administrative construction” of an ambiguous provision. C OLO . R EV . S TAT . § 2-4-203(1)(f). Colorado courts may find an “administrative interpretation” persuasive so long as it is reasonable and “consistent with public policy.” City of Aurora, 919 P.2d at 203. This is particularly so when the agency’s interpretation “reflects a long-standing construction.” Lucero v. Climax Molybdenum Co., 732 P.2d 642, 646 (Colo. 1987).

18 See Aplt.’s App. at 293 (noting the Division’s interpretation could be persuasive authority); see also Ward v. Allstate Ins. Co., 45 F.3d 353, 356 (10th Cir. 1994) (explaining that under Colorado law, even when deference is not appropriate, a state “agency interpretation” may “be treated as persuasive authority”); City of Aurora v. Bd. of Cty. Comm’rs, 919 P.2d 198, 203 (Colo. 1996) (finding “administrative interpretation” “persuasive”); Consol. Home Supply Ditch and Reservoir Co. v. Town of Berthoud, 896 P.2d 260, 265 (Colo. 1995) (treating state agency’s interpretation “as persuasive authority”); El Paso Cty. Bd. of Equalization v. Craddock, 850 P.2d 702, 705 (Colo. 1993) (“We find the [] Tax Administrator’s interpretation to be persuasive in deciding this case.”); cf. Arapahoe Cty. Pub. Airport Auth. v. Centennial Express Airlines, Inc., 956 P.2d 587, 592–93 (Colo. 1998) (noting that an opinion letter from a federal agency could be “persuasive authority”); Banner Advert., Inc. v. City of Boulder, 868 P.2d 1077, 1083 (Colo. 1994) (treating opinion letter from federal agency “as persuasive authority”).

64

We find the Division’s interpretation persuasive. As explained above, the companionship exemption is ambiguous: one could read it as applying to all companions, or as applying to only those companions employed by households or family members. Both readings are reasonable and consistent with Colorado public policy. That said, ever since 1998, the Division has consistently and repeatedly adopted the former reading. It made that point clear in the two opinion letters, and it put that interpretation into practice in the two administrative proceedings. The Division’s unbroken, oft-expressed position is potent evidence that it intended the companionship exemption to apply to companions employed by third-party employers.

The district court disagreed. It was unpersuaded by the Division’s interpretation, as expressed in its opinion letters. As noted supra, in those letters, the Division explained that its interpretation of the Wage Order’s companionship exemption as extending to companions employed by third-party employers “was intended to mirror” the corresponding federal “companions definition and associated regulations” promulgated by the U.S. Department of Labor under the FLSA, which likewise extended the exemption to companions employed by third- party employers. Id. at 189–90, 192–93. The district court rejected this position, asserting that it was contrary to the companionship exemption’s plain meaning,

65

which dictates that the exemption does not extend to companions employed by third-party employers. See Aplt.’s App. at 289–90. And because the Division’s interpretation was inconsistent with that plain meaning, the district court held that the Division’s interpretation was not entitled to deference. In support of this conclusion, the district reasoned that the Division’s letters failed to make clear whether “the [Division] was even interpreting the CWA’s [i.e., Colorado Wage Act’s] companion exemption,” seeing as those letters did not “reference[], much less analyze[]” the pertinent “CWA provisions, regulations or guidelines.” Id. at 292. Instead, the district court observed, the Division’s letters referred exclusively to the FLSA provisions related to the federal companionship exemption—despite the fact that “the FLSA and the CWA differ significantly in this regard.” In this vein, as the court read them, the letters “contain[ed] no analysis, no reasoning and not even an acknowledgement of the differences between the two laws.” Id. As such, the court mused that “it is not even clear that [the then-director of the Division] was aware the state and federal laws differed.” Id. at 292–93. The court also highlighted the disclaimer in both letters advising that “[t]he position of the [D]ivision may [] change over time,” as well as that “[t]his position is not legal advice, and the opinion of attorneys and the judicial system may differ.” Id. at 293 n.5 (quoting Aplt.’s App. at 190, 193).

66

In our view, however, the district court’s reasons for ignoring the opinion letters are mistaken. For one, the CWA does not have a companionship exemption. See C OLO . R EV . S TAT . §§ 8-4-101 to 8-4-123. Rather, the companionship exemption is contained in the Wage Order, which is a regulation—not a statute as the CWA is. For purposes of further analysis, we will assume that the district court inadvertently used the term “CWA” to mean the Wage Order. But see Aplt.’s App. at 292 (referring to “the CWA’s exemption and related regulations”). Even so, its statement that the letters did not interpret the companionship exemption is, respectfully, simply not right: the opinion letters refer to and cite the Wage Order by name. See id. at 189–90, 192–93. The district court likewise erred in stating that the Division was not “aware [that] the state and federal laws differed.” Id. at 293. Each letter explains that unlike federal law, “the Wage Order itself does not elaborate upon what constitutes the definition of a companion” or “provide specific guidance on the proper classification of a companion.” Id. at 189, 192. And the letters read like a syllogism: the Division intended the companionship exemption to mirror federal law; federal law expressly included companions employed by third-party employers within its exemption; therefore, the companionship exemption applies to companions employed by third-party employers. We are thus constrained to

67

disagree with the district court’s assessment that the letters contained “no analysis” and “no reasoning.” Id. at 292.

We also do not agree with the district court’s reliance in its analysis on the disclaimers included in the Division’s respective letters. It is true that both letters include nearly identical disclaimers, 19 which as noted supra read in relevant part as follows: “The position of the Division may . . . change over time. This position is not legal advice, and the opinion of attorneys and the judicial system may differ. The information contained herein has no bearing on the applicability or interpretation of any law or regulation not specifically addressed in the letter.” Aplt.’s App. at 193; see also id. at 190. In effect, the disclaimers clarify that the opinion letters are informal and nonbinding. But even informal and nonbinding letters “can be used as persuasive authority.” Banner Advert., Inc., 868 P.2d at 1083 (finding an informal letter from federal agency persuasive). Indeed, the Colorado Supreme Court has relied on an “Interpretive Bulletin” as persuasive authority on the meaning of a statute even though the bulletin disclaimed that the opinions therein did “not have the force and effect of rule.” Williams v. Kunau, 147 P.3d 33, 39, 41 (Colo. 2006) (quoting Colo. Dep’t of Labor & Emp’t, Div. of. Workers’ Comp., Interpretive Bulletin 11A (Mar. 6, 2006)). What is more, the

19 On appeal, Maxim claims that only the 2006 letter, but not the 2012 letter, had a disclaimer. Maxim is wrong. Both letters have identical disclaimers. See Aplt.’s App. at 190, 193.

68

opinion letters specifically address the companionship exemption. Thus, the disclaimers’ warning that the letters have “no bearing on the applicability or interpretation of any law or regulation not specifically addressed in the letter” implies that we should read the letters as indicative of the Division’s interpretation of the companionship exemption. Therefore, the district court was incorrect in using the disclaimers to justify giving the opinion letters no persuasive value.

In short, we find the Division’s longstanding and consistent interpretation of the companionship exemption as applying to companions employed by third- party employers persuasive. This persuasive authority gives us greater confidence in our conclusion—reached through our analysis of the ordinary and particular meanings of the relevant terms and application of other aids of construction—that Maxim’s reading of the companionship exemption is the superior one. We therefore hold that the companionship exemption applies to companions employed by third-party employers.

IV

To conclude, as companions employed by a third-party employer, Ms.

Jordan and her fellow class members fall within the companionship exemption. As a result, they are not entitled to overtime wages under Colorado law. For that reason, we REVERSE the district court’s final judgment, VACATE the award of

69

$2,015,253.42 in overtime wages and $691,474.20 in prejudgment interest, and REMAND the case to the district court with instructions to enter judgment in Maxim’s favor.