Jenkins v. Jordan Valley Water Conservancy District

2012 UT App 1

• Court: Court of Appeals of Utah
• Decided: January 6, 2012
• Docket: 20100400-CA
• Status: Published

Opinion

IN THE UTAH COURT OF APPEALS

‐‐‐‐ooOoo‐‐‐‐

Alan Jenkins, Ash Jenkins, and Patricia ) OPINION Jenkins, ) ) Case No. 20100400‐CA Plaintiffs and Appellants, ) ) FILED v. ) (January 6, 2012) ) Jordan Valley Water Conservancy ) 2012 UT App 1 District, ) ) Defendant and Appellee. )

‐‐‐‐‐

Third District, Salt Lake Department, 070908316 The Honorable Judith S. Atherton

Attorneys: Carl E. Kingston, Salt Lake City, for Appellants David C. Richards and Sarah E. Spencer, Salt Lake City, for Appellee

Before Judges McHugh, Voros, and Orme.

McHUGH, Presiding Judge:

¶1 Alan, Ash, and Patricia Jenkins (the Jenkinses) appeal from the trial court’s order granting summary judgment in favor of Jordan Valley Water Conservancy District (the District) on the Jenkinses’ claim for damages to their home and property after a water line owned and operated by the District broke and flooded the Jenkinses’ home on two separate occasions. The trial court concluded that the Jenkinses’ claims were barred by the public duty doctrine. We conclude that the trial court was incorrect and reverse on this issue. In addition, we reject the District’s alternative grounds for summary judgment. Although the District’s actions fall within the scope of the statutory immunity afforded governmental entities, the current definition of “governmental function” results in a complete abrogation of the Jenkinses’ preexisting remedy and violates the Utah Constitution’s open courts clause. Therefore, we remand to the trial court for further proceedings consistent with this decision.

BACKGROUND

¶2 Alan Jenkins is the owner of a home located in South Salt Lake, Utah (the Home). During the relevant time period, Alan’s son and daughter‐in‐law, Ash and Patricia Jenkins, and Ash’s and Patricia’s three children lived with Alan in the Home. The District is a political subdivision of the State of Utah. It treats and delivers water to other local districts, cities, residents, and businesses located within the geographic boundaries of the District. To accomplish these purposes, the District owns and operates approximately 275 miles of water transmission and distribution pipelines, including a section that is buried along the south side of 3300 South Street between 200 and 500 East (the Water Line Section). The Water Line Section runs parallel to the Jenkinses’ Home, which abuts 3300 South.

¶3 In December 2002, after considering numerous factors, including the history of prior breaks, the Engineering Department Manager and Distribution & Transmission Technical Advisor (collectively, the Engineers) identified the Water Line Section as needing to be replaced. Most of the pipe comprising the water line is cast iron pipe that was installed in the 1950s and 1960s; the Water Line Section was installed in 1957. These older pipes have a useful life expectancy of forty to seventy years depending upon various factors, including the depth of the pipe and soil conditions. Consequently, many sections of the water line were coming to the end of their useful life, and a significant number of sections had been identified for replacement (the Identified Pipe).

¶4 According to the District, it is financially impossible to replace all of the Identified Pipe in a single year. Therefore, the District has adopted procedures to prioritize the replacement of Identified Pipe according to various factors affecting need, cost, and convenience, and to recommend which Identified Pipe should be replaced in any given year (the Recommended Pipe). Based on the application of those factors, the

20100400‐CA 2 Engineers determined that the replacement of the Water Line Section should be deferred because it was not economically justified at that time. In their 2003‐2004 annual budget proposal to the District’s Board of Trustees (the Board), the Engineers presented their recommendation of which Identified Pipe should be replaced that year, the reasons for that recommendation, and a request for the necessary funds to replace the Recommended Pipe. The Engineers acknowledge that their decision was influenced by the Board’s expectation that they not exceed the prior year’s capital improvement budget by more than a certain percentage. Although they identified the Water Line Section as needing to be replaced in 2002, the Engineers did not recommend it for replacement until 2006. The Board allocated the necessary funds for replacement of the Water Line Section in the District’s 2006‐2007 fiscal year budget.

¶5 On November 19, 2005, after it had been listed as Identified Pipe but before the Engineers had designated it as a Recommended Pipe, the Water Line Section ruptured, resulting in the flooding of the Jenkinses’ property and the Home (the 2005 Breach). The Jenkinses contend that the 2005 Breach caused “structural damage to the [Home] and realty and destroy[ed] personal property belonging to [them].” Subsequently, the Board accepted the Engineers’ recommendation that the Water Line Section be replaced and allocated the necessary funds to do so. In October 2006, the District began replacing the six‐inch cast iron pipe in the Water Line Section with six‐inch PVC pipe. The replacement was timed to coincide with a South Salt Lake City construction project involving the sidewalk, curb, and gutter located above the Water Line Section. On October 2, 2006, while the District was laying the new PVC pipe alongside the cast iron pipe, the Water Line Section broke at a different point in the same general location as the 2005 Breach, again flooding the Home and causing additional damage (the 2006 Breach).1 The Water Line Section had broken at least nine times in the nine years preceding the Breaches, and seven of those breaches had occurred during the five years immediately preceding the 2005 Breach.

¶6 Although the District provided some financial assistance to the Jenkinses after the 2005 Breach, the Jenkinses claim that it did not compensate them fully for their damages then and that the District has refused to pay any amounts in connection with the 2006 Breach. Consequently, on November 1, 2006, the Jenkinses served a notice of

1 The 2005 Breach and the 2006 Breach are collectively referred to as “the Breaches.”

20100400‐CA 3 claim indicating their intent to sue the District. When the District failed to respond, the Jenkinses filed a complaint, seeking damages for property damage, emotional distress, and lost wages. The District answered the complaint, raising both the public duty doctrine and governmental immunity as defenses. Thereafter, the District filed a motion and memorandum for summary judgment asserting (1) that the Jenkinses’ claims were barred by the public duty doctrine; (2) that even if not barred by that doctrine, the Jenkinses could not prevail because they had failed to designate an expert; (3) that even if the Jenkinses could establish liability without an expert, the District was immune from suit; and (4) that the trial court lacked jurisdiction over some of the Jenkinses’ claims because they were not identified in the notice of claim. After full briefing and hearing, the district court granted summary judgment on the basis that the claims against the District were barred by the public duty doctrine. Because the district court found this doctrine determinative, it did not consider the other grounds for summary judgment advanced by the District. This appeal followed.

ISSUE AND STANDARD OF REVIEW

¶7 The Jenkinses appeal the trial court’s order granting summary judgment in favor of the District. “We review summary judgments for correctness, giving no deference to the trial court’s decision (even on questions that would be denominated as ‘mixed’ [questions of law and fact] if they arose on appeal after trial).” Bahr v. Imus, 2011 UT 19, ¶ 16, 250 P.3d 56.

ANALYSIS

¶8 We begin our analysis with the examination of the District’s challenge to the sufficiency of the notice of claim because we generally consider issues affecting subject matter jurisdiction first. See Kilpatrick v. Bullough Abatement, Inc., 2008 UT 82, ¶ 9, 199 P.3d 957 (“The first issue is whether we have jurisdiction . . . .”); Hicks v. UBS Fin. Servs., Inc., 2010 UT App 26, ¶ 12, 226 P.3d 762 (“[B]efore reaching the substantive issue . . . we must first determine whether we have jurisdiction to hear this appeal.”), cert. granted, 238 P.3d 443 (Utah 2010) (No. 20100186). We conclude that subject matter jurisdiction exists with respect to the Jenkinses’ claims, but we remand to the trial court for the resolution of factual issues determinative of the damages available to the Jenkinses. We

20100400‐CA 4 then proceed to the issue of whether the Jenkinses’ negligence claim is barred by the public duty doctrine. See generally Rollins v. Petersen, 813 P.2d 1156, 1162 n.3 (Utah 1991) (noting that the public duty doctrine should be considered before questions of immunity). We conclude that it is not and reverse the trial court’s summary judgment in favor of the District on that theory.

¶9 We next address the alternative grounds raised by the District in support of summary judgment in its favor. Because “judicial restraint requires that courts avoid reaching constitutional questions in advance of the necessity of deciding them,” State v. Thurman, 846 P.2d 1256, 1262 (Utah 1993) (internal quotation marks omitted), we defer the open courts challenge advanced by the Jenkinses until after the consideration of all other issues. See Clegg v. Wasatch Cnty., 2010 UT 5, ¶¶ 26‐27, 227 P.3d 1243 (declining to reach an open courts clause question when a factual issue remained with respect to the application of immunity because “to do so would be to impermissibly render an advisory opinion”). Thus, we begin our analysis of the District’s alternative grounds by considering whether summary judgment is appropriate because the Jenkinses failed to designate an expert. Under the unique facts present here, we hold that expert testimony is not required. We then proceed to the issue of whether, even if the District negligently caused damage to the Jenkinses, it is immune from liability under the Governmental Immunity Act of Utah (the GIAU). See Utah Code Ann. §§ 63G‐7‐101 to ‐904 (2011). Ultimately, we conclude that the District’s decision of when to replace the Water Line Section was a discretionary function for which the District is immune as a matter of law.

¶10 Having thus eliminated all other bases for resolving the issues before us, we proceed to the Jenkinses’ argument that even if the District would be immune, summary judgment is not appropriate because the GIAU as applied to their claims violates the open courts clause of the Utah Constitution. See Utah Const. art. I, § 11; see also Day v. State, 1999 UT 46, ¶ 9, 980 P.2d 1171 (following “the principle that a case should be decided on non‐constitutional grounds if possible and that constitutional issues should be addressed only when necessary”). We first determine that legislative amendments to the state’s governmental immunity laws resulted in the complete abrogation of the Jenkinses’ preexisting remedy without providing them a reasonable alternative. We next focus on the social or economic evils identified by the Utah Legislature as the motivation for the amendments, deferring to that body’s judgment that these concerns are legitimate and require a remedy. Upon the identification of the evils to be addressed, we proceed to the consideration of whether the selected remedy is

20100400‐CA 5 reasonable and rational, including the review of whether the remedy is narrowly tailored. Although we are convinced that the means selected by the legislature are rationally designed to address the economic and social evils it has identified, we conclude that the complete abrogation of the Jenkinses’ claims is not narrowly tailored. Based on controlling precedent from the Utah Supreme Court, we hold that the open courts clause of the Utah Constitution prevents the District from taking advantage of the immunity afforded by the GIAU with respect to the Jenkinses’ claims. Therefore, we reverse the decision of the trial court and remand for further proceedings consistent with this decision.

I. Subject Matter Jurisdiction

¶11 We first consider the District’s challenge to our subject matter jurisdiction over the Jenkinses’ second and third causes of action on the ground that they were not included in the notice of claim as required by the GIAU. See Utah Code Ann. § 63G‐7‐ 401(3)(a) (2011) (requiring that any person with a claim against a governmental entity “file a written notice of claim with the entity before maintaining an action”);2 Greene v. Utah Transit Auth., 2001 UT 109, ¶ 16, 37 P.3d 1156 (“Compliance with the Immunity Act is necessary to confer subject matter jurisdiction upon a trial court to hear claims against governmental entities.”).3 Whether this court has subject matter jurisdiction is a question of law that we review under the correction of error standard, affording no deference to the trial court’s legal conclusion. See Xiao Yang Li v. University of Utah, 2006

2 Since the Jenkinses filed their notice of claim on November 1, 2006, section 63G‐ 7‐401 has been amended several times. See Utah Code Ann. § 63‐30d‐401(3)(a) (2004); id. § 63G‐7‐401(3)(a) (2008); id. § 63G‐7‐401(3)(a) (2011). Because those amendments did not affect the provisions regarding the content of the notice of claim, we refer to the current version of the statute. Compare id. § 63‐30d‐401(3)(a) (2004), with id. § 63G‐7‐ 401(3)(a) (2011). 3 The parties agree that the District is a governmental entity. See id. § 63G‐7‐102(3) (2011) (“‘Governmental entity’ means the state and its political subdivisions as both are defined in this section.”); id. § 63G‐7‐102(7) (“‘Political subdivision’ means any county, city, town, school district, community development and renewal agency, special improvement or taxing district, local district, special service district, an entity created by an interlocal agreement . . . , or other governmental subdivision or public corporation.”).

20100400‐CA 6 UT 57, ¶ 7, 144 P.3d 1142. The interpretation of the GIAU is also a question of law that we review under the nondeferential standard of correctness. See General Constr. & Dev., Inc. v. Peterson Plumbing Supply, 2011 UT 1, ¶ 5, 248 P.3d 972.

¶12 The GIAU mandates that before a party may commence an action against a governmental entity in the district court the party must file a written notice of claim that includes, to the extent relevant here, “(i) a brief statement of the facts; (ii) the nature of the claim asserted; [and] (iii) the damages incurred by the claimant so far as they are known.” See Utah Code Ann. § 63G‐7‐401(3)(a). “The purpose of the notice of claim requirement is to provide the governmental entity an opportunity to correct the condition that caused the injury, evaluate the claim, and perhaps settle the matter without the expense of litigation.” Mecham v. Frazier, 2008 UT 60, ¶ 17, 193 P.3d 630 (internal quotation marks omitted). Thus, the issue presented by the District is whether the Jenkinses’ notice meets the requirements of section 63G‐7‐401.

¶13 The Jenkinses’ notice of claim states, in relevant part,

On or about November 19, 2005, and again on or about October 2, 2006, the water line running along 3300 South Street broke, flooding the Jenkin[se]s’ [H]ome, causing extensive damage to the foundation and walls of the [H]ome, damaging or destroying the personal property of the occupants and creating an environment where mold is now present throughout the basement of the [H]ome.

. . . [T]he damage . . . to the foundation of the house has created an unsafe condition and . . . to remedy the condition, the foundation may have to be replaced. The cost of repairing or replacing the foundation is not yet known, but . . . [t]he rough estimate is in the neighborhood of $50,000. Mr. Jenkins has received an estimate to clean up the mold from Watertite Solutions, in the amount of $28,716.60. The damage to the personal property damaged or destroyed is approximately $15,000.00.

20100400‐CA 7 When the Jenkinses received no response to their notice, they filed their complaint, seeking compensation for the damage to the Home and their personal property. In addition, they sought lost wages and damages for the “extreme emotional anguish and suffering” all of the Jenkinses have endured as a result of the living conditions created by the flooding, including persistent mold, a damaged and now‐porous foundation, an inoperable water heater, and an inoperable heating and air conditioning system. According to the District, the notice of claim did not fairly apprise the District of the additional claims for lost wages and emotional distress.

¶14 Although the GIAU has been interpreted to require strict compliance, the court will not impose obligations that are not included in the statute. See Peeples v. State, 2004 UT App 328, ¶ 9, 100 P.3d 254 (“Strict compliance [with section 63G‐7‐401] is not . . . a one‐way street, and a claimant is not required to do more than the [GIAU] clearly requires.”). Further, the statute does not mandate that the notice of claim “meet the standards required to state a claim for relief,” as in a complaint filed to initiate a civil action. See Houghton v. Department of Health, 2005 UT 63, ¶ 21, 125 P.3d 860 (internal quotation marks omitted). “Rather, a plaintiff need only include enough specificity in the notice to inform as to the nature of the claim so that the defendant can appraise its potential liability.” Id. (internal quotation marks omitted); see also id. ¶ 22 (holding that the draft complaint requesting “attorney[] fees as may be provided by law,” attached to the notice of claim fairly apprised the governmental entity that the plaintiffs would seek attorney fees later requested under a particular statute (internal quotation marks omitted)); Mecham, 2008 UT 60, ¶ 19 (holding that a notice of claim against a governmental employee need not use words such as “malice” or “fraud,” so long as the notice alerts the governmental entity of an intent to sue due to the employee’s fraudulent or malicious conduct).

¶15 In interpreting the plain language of the GIAU, this court has concluded that the requirement that the plaintiff provide a “brief statement of the facts” unambiguously “does not require specifics.” See Peeples, 2004 UT App 328, ¶ 8; see also Utah Code Ann. § 63G‐7‐401(3)(a)(i). Here, the Jenkinses’ statement “contains multiple facts, including the date of [their] injury, its alleged cause, details of the alleged . . . defect, and that the injury occurred at [the Home].” See Peeples, 2004 UT App 328, ¶ 10. Although it does not include any facts concerning the impact of the flooding on the emotional well being or earning capacity of the Jenkinses, the notice adequately “constitutes a brief recitation of those facts most relevant to the [incident].” See Xiao Yang Li, 2006 UT 57, ¶ 10.

20100400‐CA 8 ¶16 The statute also requires the claimant to set forth “the nature of the claim.” See Utah Code Ann. § 63G‐7‐401(3)(a)(ii) (2011). “There is no ambiguity in the nature of claim requirement: ‘There must be enough specificity in the notice to inform as to the nature of the claim so that the defendant can appraise its potential liability.’” Heideman v. Washington City, 2007 UT App 11, ¶ 14, 155 P.3d 900 (quoting Yearsley v. Jensen, 798 P.2d 1127, 1129 (Utah 1990)). While the notice of claim filed by the Jenkinses alerts the District to its potential liability for the substantial damages to the Jenkinses’ Home and property, it does not specifically notify the District that they also hope to recover for emotional distress and lost wages. The District asserts that this renders the notice of claim invalid with respect to those claims.

¶17 In Yearsley v. Jensen, 798 P.2d 1127 (Utah 1990), the Utah Supreme Court interpreted the “nature of the claim” requirement. See id. at 1129. There, the plaintiff filed a notice of claim seeking “$100,000 for physical and emotional distress” suffered when police officers “physically beat the claimant in the course of the arrest of a third party.” Id. at 1128. When the plaintiff sued for malicious prosecution, the defendants argued that the notice of claim had not adequately alerted them to that cause of action. See id. The Utah Supreme Court agreed, concluding that the claim for malicious prosecution “var[ied] so profoundly” from the plaintiff’s notice of claim that it would do violence to the immunity act to permit the plaintiff to go forward with that claim in the litigation. See id. at 1129; see also Heideman, 2007 UT App 11, ¶ 14 (holding that the court was judicially barred from considering plaintiff’s claim for intentional interference with economic relations where the notice of claim sought damages for breach of contract, civil rights violations, and other unrelated causes of action). The District contends here that, as in Yearsley, the Jenkinses’ claims for lost wages and emotional distress vary profoundly from the nature of the claim described in their notice of claim. See Yearsley, 798 P.2d at 1129.

¶18 However, the Jenkinses have not asserted any new causes of action; instead, they seek categories of damages not previously identified.4 In Behrens v. Raleigh Hills Hospital Inc., 675 P.2d 1179 (Utah 1983), the Utah Supreme Court considered that distinction

4 Although the Jenkinses’ complaint asserts the claims for lost wages and emotional distress as separate causes of action, they incorporate by reference the allegations concerning the District’s failure to replace the Water Line Section from the first cause of action as the basis of liability for those additional damages.

20100400‐CA 9 determinative. See id. at 1182. There, the defendant argued that the plaintiff could not recover punitive damages because they had not been identified in the notice of claim. Id. The supreme court disagreed, explaining that “an amendment to include [punitive] damages does not import into a case a new and different cause of action.” See id. Similarly, the Jenkinses’ request to be compensated for lost wages and emotional distress caused by the flooding does not import a new cause of action into the case. And despite the absence of a claim for those damages, the notice of claim is detailed enough to provide the District an opportunity to “correct the condition that caused the injury, evaluate the claim, and perhaps settle the matter without the expense of litigation.” Mecham v. Frazier, 2008 UT 60, ¶ 17, 193 P.3d 630 (internal quotation marks omitted).

¶19 The District’s argument seems more accurately characterized as a challenge to the Jenkinses’ compliance with the requirement that the notice of claim include a description of “the damages incurred by the claimant so far as they are known.” See Utah Code Ann. § 63G‐7‐401(3)(a)(iii). Our supreme court reviewed the scope of that requirement in Xiao Yang Li v. University of Utah, 2006 UT 57, 144 P.3d 1142, where it considered whether a complaint filed by the heirs of persons killed in a traffic accident should have been dismissed. See id. ¶¶ 2‐5. The heirs filed a notice of claim describing the accident and indicating that the nature of the claim was negligence, negligent supervision, and vicarious liability. See id. ¶¶ 10‐11. Although the notice stated only that seven of the plaintiffs were dead and that the others were receiving medical treatment, the supreme court concluded that it “went beyond the minimal requirements of the [GIAU] by at least indicating that the damages would be those incurred as a result of the injuries and deaths caused by the accident.” See id. ¶ 12. The supreme court further explained that the GIAU “does not call for an all‐inclusive summary of the damages or for speculation.” See id.

¶20 Here, the Jenkinses put the District on notice of their intent to sue for the damages incurred as a result of the Breaches. They went beyond the minimal requirements of the GIAU by quantifying many of those damages. While they did not provide an all‐inclusive summary of damages, that omission does not necessarily render their notice of claim inadequate. The GIAU requires that the notice of claim include a description of the damages incurred “so far as they are known.” See Utah Code Ann. § 63G‐7‐401(3)(a)(iii). The Jenkinses filed their notice of claim one month after the 2006 Breach of the Water Line Section and included significant detail in quantifying the damages then known. While the notice of claim does not include lost

20100400‐CA 10 wages and emotional distress damages, it was written before the Jenkinses were forced to live in the damaged Home for years pending resolution of this dispute. Indeed, the complaint alleges that the emotional distress is continuing: “The[] conditions [to the Home caused by the flooding] continue to cause extreme mental anguish and suffering to the [Jenkinses], to their damage in an amount to be proved at trial.”

¶21 The extent to which the damages for the emotional impact of the flooding arose after the Jenkinses filed their notice of claim is not apparent from the record. Likewise, the claim for lost wages does not provide any information about when “Ash Jenkins and [Patricia] Jenkins were required to take time off from their usual vocations.” As a result, we are unable to determine whether the Jenkinses met their obligation to provide a “listing of only known damages.” See Xiao Yang Li, 2006 UT 57, ¶ 12. If the damages were known to the Jenkinses but not reported in the notice of intent, recovery of them is barred; to the extent that the Jenkinses were not aware of these damages at the time they filed the notice of claim, recovery is not barred. Further proceedings in the trial court are necessary to resolve this factual issue.

¶22 In any event, this court has subject matter jurisdiction over the claims and damages that were described in the notice of claim. Therefore, we now proceed to the review of the substantive issues raised by this appeal, beginning with the basis for the trial court’s decision.

II. The Public Duty Doctrine

¶23 The Jenkinses contend that the trial court erred in concluding that the public duty doctrine bars their claim that the District was negligent in failing to replace the Water Line Section prior to the Breaches.5 The trial court determined that under the public duty doctrine, the District owed them no duty to prevent the Breaches. Whether a duty exists is a question of law that we review for correctness. See Slisze v. Stanley‐ Bostitch, 1999 UT 20, ¶ 9, 979 P.2d 317; Tuttle v. Olds, 2007 UT App 10, ¶¶ 20‐24, 155 P.3d

5 The Jenkinses clarified at oral argument that they do not assert any negligence in connection with the repair of the 2005 Breach or the replacement of the Water Line Section in 2006. Instead, they claim that the District negligently failed to replace the cast iron pipe comprising the Water Line Section before the Breaches, thereby causing extensive damage to the Jenkinses.

20100400‐CA 11 893 (McHugh, J., concurring in part and dissenting in part) (reviewing the trial court’s reliance on the public duty doctrine to grant summary judgment as a question of law).

A. The Relationship Between the Public Duty Doctrine and Negligence

¶24 The Jenkinses’ claim is based on a theory of negligence. To prove a claim of negligence, the Jenkinses must establish four elements: “(1) that the defendant owed the plaintiff a duty, (2) that the defendant breached that duty, (3) that the breach of the duty was the proximate cause of the plaintiff’s injury, and (4) that the plaintiff in fact suffered injuries or damages.” Webb v. University of Utah, 2005 UT 80, ¶ 9, 125 P.3d 906 (internal quotation marks omitted). The failure to prove any one of these elements is fatal to the plaintiff’s case. See Dikeou v. Osborn, 881 P.2d 943, 946 (Utah Ct. App. 1994). The issue before us is whether the trial court correctly concluded that the District owed no duty to the Jenkinses, thereby defeating their claim for negligence as a matter of law.

¶25 “‘“[D]uty” is a question of whether the defendant is under any obligation for the benefit of the particular plaintiff; and in negligence cases, the duty is always the same—to conform to the legal standard of reasonable conduct in light of the apparent risk.’” Downing v. Hyland Pharmacy, 2008 UT 65, ¶ 11, 194 P.3d 944 (quoting W. Page Keeton et al., Prosser and Keeton on the Law of Torts § 53 (5th ed., Lawyer’s Ed. 1984)). Typically, the court determines whether a duty exists based upon “the sum total of those considerations of policy which lead the law to say that the particular plaintiff is entitled to protection.” See DeBry v. Valley Mortg. Co., 835 P.2d 1000, 1003‐04 (Utah Ct. App. 1992); accord Webb, 2005 UT 80, ¶ 9. The public duty doctrine supplants this analysis by predetermining that policy considerations weigh against imposing a duty on the governmental entity to protect any particular plaintiff. It is

a common law rule[ that] provides that a governmental entity and its employees owe no duty of care to individual members of the general public to provide governmental services and this rule of nonliability is grounded in the principle that the duty of a governmental entity to preserve the well‐being of the community is owed to the public at large, rather than to specific members of the community.

57 Am. Jur. 2d Municipal, County, School, and State Tort Liability § 88 (2001). Thus,

20100400‐CA 12 “[f]or a governmental agency and its agents to be liable for negligently caused injury suffered by a member of the public, the plaintiff must show a breach of a duty owed him as an individual, not merely the breach of an obligation owed to the general public at large by the governmental official.”

Day v. State, 1999 UT 46, ¶ 11, 980 P.2d 1171 (quoting Ferree v. State, 784 P.2d 149, 151 (Utah 1989)).

B. The Distinction Between the Public Duty Doctrine and Governmental Immunity

¶26 In contrast to the public duty doctrine, governmental immunity provides that a governmental entity need not pay any damages, despite the fact that it was negligent. Thus, immunity is analytically distinct from the public duty doctrine. Nevertheless, the simultaneous operation of the two legal theories has caused considerable confusion.6 See Webb, 2005 UT 80, ¶ 7 (“The central challenge confronting us in this case is to make sense of the scene where common law negligence and governmental immunity law have collided.”). Because governmental immunity “is an affirmative defense and conceptually arises subsequent to the question of whether there is tort liability in the first instance,” see Ferree, 784 P.2d at 153, the “proper mode of analysis is to first consider whether there is a legal theory upon which suit can be brought . . . before

6 Due to the confusion caused by this overlap, some states have abolished the common law public duty doctrine. See, e.g., Natrona Cnty. v. Blake, 2003 WY 170, ¶ 15, 81 P.3d 948, 956 (Wyo. 2003) (collecting cases abolishing the public duty doctrine); 1 Dan B. Dobbs, The Law of Torts § 271 (2001) (noting the rejection of the doctrine in some states); 18 McQuillin Municipal Corporations § 53.04.25 (3d. ed. 2003) (collecting cases). Indeed, Chief Justice Durham has acknowledged that the interplay between governmental immunity and public duty in Utah creates “confusion in the law and inequitable results.” See Rollins v. Petersen, 813 P.2d 1156, 1166 (Utah 1991) (Durham, J., concurring and dissenting). However, our supreme court has declined to eliminate the common law doctrine in response to the legislature’s adoption of statutory governmental immunity. See id. at 1162 n.3 (majority opinion) (“[T]he legislature’s abrogation of absolute sovereign immunity does not lead to the conclusion that the public duty doctrine has also been abrogated.”).

20100400‐CA 13 considering the separate and independent questions of whether the [entity] is immune.” Rollins v. Petersen, 813 P.2d 1156, 1168 n.3 (Utah 1991); see also Ledfords v. Emery Cnty. Sch. Dist., 849 P.2d 1162, 1163‐64 (Utah 1993) (noting that analytical clarity is served by addressing traditional tort questions before issues of immunity). We follow that mode of analysis now by first reviewing whether, under the public duty doctrine, the District owed no duty of care to the Jenkinses individually.

C. The Public Duty Doctrine in Utah

¶27 The Utah Supreme Court first relied on the public duty doctrine in Obray v. Malmberg, 26 Utah 2d 17, 484 P.2d 160 (1971), where it cited cases from other jurisdictions to support its conclusion that the “failure by a public sheriff to investigate a crime . . . [is] not pursuable by an individual since the public official’s duty is to the public.”7 See id. at 1162. Since then, the public duty doctrine has been applied in Utah to eliminate a duty to any individual member of the public for the negligent performance of a governmental regulatory or enforcement activity. See, e.g., Madsen v. Borthick, 850 P.2d 442, 446‐47 (Utah 1993) (holding that the Utah Department of Finance had no duty to plaintiffs in the regulation of a financial institution); C.T. v. Martinez, 845 P.2d 246, 247 (Utah 1992) (holding that the Utah Department of Social Services had no duty to plaintiffs in licensing a day care facility); Ferree, 784 P.2d at 152 (holding that the Utah State Department of Corrections did not owe a duty to plaintiff in granting a prisoner a two‐day release). Indeed, the application of the doctrine to these “[v]arious kinds of regulatory and police activities” is a classic form of the public duty doctrine. See DeBry, 889 P.2d at 440 n.12; see also 57 Am. Jur. 2d Municipal, County, School, and State Tort Liability § 102 (2001) (“[A] governmental entity cannot be held liable in negligence for failure to carry out its statutory duties.”); 1 Dan B. Dobbs, The Law of Torts § 271 (2001) (“The traditional examples of the public duty rule are all cases of non‐ performance of a statutory duty, not cases of affirmative conduct.”). In addition, Utah’s public duty doctrine provides protection to governmental entities engaged in activities that benefit the public at large, whether or not considered regulatory or enforcement

7 The doctrine was first recognized in the United States in South v. Maryland, 59 U.S. 396 (1855). There, the Supreme Court held that a sheriff’s common law duty to enforce the law did not create a duty to the plaintiff that was then breached when the sheriff failed to intervene after the plaintiff was kidnapped and held for ransom. See id. at 402‐03.

20100400‐CA 14 activities. See, e.g., Cannon v. University of Utah, 866 P.2d 586, 589 n.6 (Utah Ct. App. 1993) (holding that the police officers posted by the university at a crosswalk in which the plaintiffs were hit on their way to a university‐sponsored basketball game owed “a general duty . . . to the public at large, not to any distinct group”).

D. Application of the Public Duty Doctrine to These Facts

¶28 The Jenkinses argue that the public duty doctrine does not apply to the District’s actions here because the sale of water is a proprietary, rather than a governmental, function. See 57 Am. Jur. 2d Municipal, County, School, and State Tort Liability § 102 (“The public duty rule applies to uniquely government functions.”).8 However, in Utah, the public duty doctrine is applied more broadly and arguably includes proprietary functions of governmental entities. See, e.g., Webb, 2005 UT 80, ¶ 28 (holding that the university had no duty to a student injured on an icy sidewalk during a required, school‐related field trip); Rollins, 813 P.2d at 1161‐62 (holding that the state‐run mental hospital had no duty to a victim injured by an escapee). Thus, whether the allegedly negligent activity is proprietary or governmental is not determinative of whether the doctrine is implicated.

¶29 Instead, the Utah Supreme Court has focused its analysis on the relationship between the plaintiff and the public entity. In Webb v. University of Utah, 2005 UT 80, 125 P.3d 906, the supreme court explained that “an omission or failure to act can generally give rise to liability only in the presence of some external circumstance—a special relationship.” Id. ¶ 10. Typically, these relationships arise “‘when one assumes responsibility for another’s safety or deprives another of his or her normal opportunities for self‐protection.’” Id. (quoting Beach v. University of Utah, 726 P.2d 413, 415 (Utah 1986)). While the supreme court acknowledged that “as a matter of public

8 The District correctly notes that the Jenkinses did not argue to the trial court that the public duty doctrine, as opposed to governmental immunity, is limited to governmental functions. However, the Jenkinses did argue that the public duty doctrine is not applicable here, citing Utah decisions where liability has been imposed. Because the applicability of the public duty doctrine turns on the relationship between the public actor and the plaintiff, rather than on whether the activity is governmental or proprietary, we conclude that the Jenkinses’ challenge to the district court’s application of the public duty doctrine was adequately preserved.

20100400‐CA 15 policy, we do not expose governmental actors to tort liability for all mishaps that may befall the public,” it held that the governmental actor will be liable when its negligence causes injury “to persons who stand so far apart from the general public that we can describe them as having a special relationship to the governmental actor.” Id. ¶ 11. In Day v. State, 1999 UT 46, 980 P.2d 1171, the supreme court provided a nonexclusive list of the circumstances under which this type of special duty can arise: (1) where there is a “statute intended to protect a specific class of person of which the plaintiff is a member from a particular type of harm”; (2) “when a government agent undertakes specific action to protect a person or property”; (3) where “government actions . . . reasonably induce detrimental reliance by a member of the public”; and (4) “under certain circumstances, when the agency has actual custody of the plaintiff or of a third person who causes harm to the plaintiff.” See id. ¶ 13; accord Webb, 2005 UT 80, ¶ 25; Gabriel v. Salt Lake City Corp., 2001 UT App 27, ¶ 17, 34 P.3d 234. The District contends that the Jenkinses do not fall within any of these categories and therefore, it owed them no duty to operate its water distribution pipeline nonnegligently. We are not persuaded.

¶30 First, the public duty doctrine in Utah has not been traditionally applied to protect a governmental entity from negligent maintenance of its own equipment when undertaken to protect an identifiable group of persons, rather than the public at large. In Nixon v. Salt Lake City Corp., 898 P.2d 265 (Utah 1995), the plaintiff, who worked for a janitorial service hired by the city to clean the Salt Lake City Airport, was injured while using a battery‐powered floor scrubber owned and maintained by the city. See id. at 267. The plaintiff presented evidence that the city had negligently maintained the equipment. See id. In response, the city argued that it was immune from suit under the governmental immunity act for any portion of its liability arising out of a negligent inspection of the equipment. See id. at 270. The Nixon court concluded that “as a matter of law, the acts complained of are acts of maintenance,” for which the city owed a duty to foreseeable plaintiffs under traditional tort principles. See id. In doing so, it relied on jurisprudence developed under the public duty doctrine. See id. at 271. The court indicated that the public duty doctrine barred recovery where an inspection was undertaken “to determine [whether] a particular building or piece of equipment was unsafe for the public as a whole,” see id., but not where the inspection was undertaken “‘to protect a particular individual or class of individuals,’” see id. at 271 (quoting 57A Am. Jur. 2d Negligence § 376 (1989)). Because the city’s inspection and maintenance of its floor scrubber was undertaken to protect persons like the plaintiff who might use the

20100400‐CA 16 equipment, the court concluded that it did not fall within the public duty doctrine or, by extension, the governmental immunity statute. See id. ¶31 This court reached a similar conclusion in Ilott v. University of Utah, 2000 UT App 286, 12 P.3d 1011, again relying on the public duty doctrine to define the limits of governmental immunity for inspections. See id. ¶¶ 12‐15. There, the plaintiff, who was attending a University of Utah football game, was injured when a plank on the bleachers broke. See id. ¶ 2. In holding that neither the public duty doctrine nor governmental immunity prevented the plaintiff from recovering, the court concluded that the University was engaged in maintaining its own property and that the plaintiff was “a member of a specific group (business invitees paying to attend a football game).” See id. ¶ 15.

¶32 In this case, we hold that the Jenkinses had a special relationship to the District and, accordingly, that the District owed them a duty of care. Unlike the public at large, the Jenkinses owned property next to the Water Line Section that the District had identified for replacement due, in part, to prior breaches. Where the District’s own criteria had identified the Water Line Section for replacement, the District “assume[d] responsibility” for its replacement and, by extension, the “safety” of individuals like the Jenkinses who lived in the path of any water that might escape from future breaches. See Beach, 726 P.2d at 415 (noting that a special relationship arises “when one assumes responsibility for another’s safety or deprives another of his or her normal opportunities for self‐protection”). In addition, after the Jenkinses were flooded in 2005 and had received some compensation from the District, they certainly stood “so far apart from the general public that we can describe them as having a special relationship to the governmental actor.” See Webb, 2005 UT 80, ¶ 11. Furthermore, the Jenkinses could not protect themselves. Thus, the District had a duty to the Jenkinses not to negligently delay the replacement of the Water Line Section. Consequently, we hold that the trial court erred when it granted summary judgment in favor of the District on the ground that the public duty doctrine absolved it of any duty to the Jenkinses to maintain the Water Line Section.9

9 In reaching this conclusion, we do not express any opinion about whether the District has a special relationship with other persons who own property adjacent to segments of the water line that have not previously breached or been identified for replacement.

20100400‐CA 17 ¶33 Nevertheless, the District contends that summary judgment in its favor can be affirmed on several alternative grounds, including that the Jenkinses failed to designate an expert and that the decision of when and whether to replace the Water Line Section was a discretionary function for which the District is immune under the GIAU. We now consider whether there are alternative grounds apparent from the record under which the trial court’s summary judgment ruling can be affirmed. See Bailey v. Bayles, 2002 UT 58, ¶ 13, 52 P.3d 1158 (holding that an appellate court may affirm the decision of the trial court on any ground apparent from the record). To affirm the summary judgment ruling on an alternative legal theory, the necessary facts supporting that theory must be undisputed. See Richardson v. Hart, 2009 UT App 387, ¶ 14, 223 P.3d 484.

III. Failure to Designate an Expert

¶34 There is no dispute that the Jenkinses did not designate an expert witness to testify as to the standard of care or whether the District breached that standard by continuing to use the Water Line Section after it was identified for replacement.10 According to the District, the lack of expert testimony on these points defeats the Jenkinses’ negligence claim because “issues of fact which are outside the knowledge and experience of lay persons must be established by expert testimony.” Hoopiiaina v. Intermountain Health Care, 740 P.2d 270, 271 (Utah Ct. App. 1987) (citing Kim v. Anderson, 610 P.2d 1270, 1271 (Utah 1980)).

¶35 The District relies on District of Columbia v. Arnold & Porter, 756 A.2d 427 (D.C. 2000), for the proposition that “the operation and maintenance of a municipal water main system and the handling of leaks in that system are not subjects within the common knowledge of jurors.” Id. at 433‐34. There, the plaintiffs owned businesses that suffered damage when the District of Columbia (D.C.) did not immediately investigate reports of a leak in its municipal water system and then delayed turning off the water in the area for approximately nine hours. See id. at 429‐32. After a bench trial on the plaintiffs’s negligence claims, the trial court found for the plaintiffs and D.C. appealed. See id. at 429. Although the District of Columbia Court of Appeals remanded on other issues, it rejected the plaintiffs’ argument that while the “general subject of municipal water standards may be beyond the ken of the average layperson,” the issues of the particular case were a matter of “common sense.” See id. at 432. The appellate

10 The parties have not raised any issues with respect to causation.

20100400‐CA 18 court first held that the operation and maintenance of the system, as well as the handling of leaks, were not subjects within the common knowledge of lay jurors, and then further explained that, given our repeated articulation of the need to show a national standard of care if the subject in question . . . is so distinctly related to some science, profession or occupation as to be beyond the ken of the average layperson, we are not persuaded . . . that the factual and legal context of [D.C.’s] actions, in conjunction with common sense make clear that no expert was required to tell the average layperson what [D.C.] should have done.

Id. at 433‐34 (citation and internal quotation marks omitted).

¶36 Based on this authority, the District asserts that

[t]he specialized knowledge regarding cast iron pipelines, including their expected and reasonably anticipated lifetimes, the nature of breaks to such pipelines, the effect of various kinds of soil conditions, the proper means of repair, the proper interpretation of a history of breaks to a particular pipeline, the identification and prioritization of [Identified Pipe], and the available and proper maintenance measures, including replacement of water lines, are all matters beyond the scope of a typical layperson’s knowledge.

Thus, it asks us to affirm the summary judgment ruling in its favor due to the Jenkinses’ failure to designate an expert on these issues.

¶37 However, unlike the plaintiffs in Arnold & Porter, who claimed that D.C. acted negligently in the manner in which it handled reports of a leak in its system, see id. at 429‐32, the Jenkinses do not claim that the District negligently investigated whether the Water Line Section needed to be replaced or that it negligently repaired the breaks in the Water Line Section. Rather, the Jenkinses contend that after making the determination that the Water Line Section needed to be replaced, the District

20100400‐CA 19 negligently waited over three years before actually replacing it, during which time the section twice ruptured and flooded their property. Specifically, the Jenkinses claim that

when the District’s own procedures established that the [Water Line Section] had failed so many times, in such a short distance, over such a short period of time, and that the District had determined three years before the Jenkin[se]s’ [H]ome was flooded that the [Water Line Section] needed to be replaced, its failure to act constituted negligence. This is particularly true in [this] case, where the District not only failed to act before the first flood, but failed to act until the Jenkin[se]s’ [H]ome had been flooded a second time, almost one year later.

Where the issue is narrowly focused on the decision to delay three years before replacing the Water Line Section, we agree with the Jenkinses that the facts and results of the District’s actions and inactions are not beyond the scope of laypersons’ knowledge and ability to analyze and determine whether the District was negligent in this case. See, e.g., Bowman v. Kalm, 2008 UT 9, ¶ 13, 179 P.3d 754 (holding that no expert was required to explain the “causal connection between a decedent made clumsy due to a doctor’s negligence, and the decedent’s death due to a dresser being pulled down on top of her”). But see Spafford v. Granite Credit Union, 2011 UT App 401, ¶¶ 26‐32 (requiring expert testimony to establish breach of duty and causation where the plaintiff tripped and fell on a curb that was alleged to be “unusually tall, ‘tapered aggressively,’ and . . . ‘in a state of disrepair’”). In response, the District contends that it was financially unable to replace all of the cast iron sections of the water line in a single year and was, therefore, required to defer replacement of the Water Line Section. We are similarly unconvinced that expert testimony is needed on that point.

¶38 In reaching this conclusion, we find this court’s decision in Schreiter v. Wasatch Manor, Inc., 871 P.2d 570 (Utah Ct. App. 1994), instructive. There, a resident of a high‐ rise apartment building, who had sustained personal injuries during a fire, brought a negligence action against the owner of the building based on its failure to install a fire sprinkler system. See id. at 572. The trial court granted summary judgment against the resident on two alternative grounds: (1) that the plaintiff failed to show the cost of or the defendant’s ability to pay for a fire sprinkler system; and (2) that the plaintiff had not designated an expert witness. See id. On appeal, this court reversed. See id.

20100400‐CA 20 ¶39 First, we held that expert testimony was not required to establish the standard of care or that it was breached. See id. at 573, 575. While the Schreiter court acknowledged that the cost of installing the fire sprinkler system might have some relevance, it held that “a jury could reasonably find that the approximate cost of such a system would not preclude its installation by a reasonable person under the circumstances.” Id. at 574. Likewise, although the cost of replacing sections of the water line may have some relevance to the question of whether the District’s delay in replacing the Water Line Section was negligent, that question “is simply not a situation where the issues or facts appear to be so complex or technical that they would otherwise elude the mental processes of the average citizen.” See id. at 575.

¶40 Next, the Schreiter court determined that the defendant’s subjective ability to pay for such a system was irrelevant to the objective standard of care to be used in determining whether the defendant had breached its duty to the resident, stating,

[The defendant’s] own actual or subjective ability to pay has no bearing at all on whether it breached its duty to [the plaintiff]. Were we to accept [the defendant’s] requirement, aside from the practical problem of requiring a plaintiff to prove a particular defendant’s financial position and capacity to pay, we would be imposing one standard of care on wealthy defendants and a lower standard on poor defendants; a higher standard on efficient and well‐managed entities and a lower standard on inefficient, poorly managed entities.

Id. at 574. The same considerations are appropriate here. The fact that a defendant operates a large quantity of negligently maintained or obsolete equipment should not affect the standard of care required to fulfill the defendant’s duty to foreseeable plaintiffs. Indeed, we have discovered nothing in our jurisprudence that would support a sliding scale of care, which benefits a defendant who operates so much outdated equipment that replacing it all at once is not feasible, while imposing a greater standard of care on the defendant who employs newer and better maintained equipment. Nor can we imagine a public policy reason for doing so. Consequently, we hold that no expert testimony was required to address whether the District had sufficient financial

20100400‐CA 21 resources to replace the Water Line Section sooner because such an opinion is irrelevant to the standard of care owed to the Jenkinses and whether that standard was breached.

¶41 In sum, we decline the District’s invitation to affirm the trial court’s summary judgment in its favor on the alternative ground that the Jenkinses failed to designate an expert witness. The issue of whether three years was a reasonable time to delay replacing the Water Line Section after the Engineers designated it as Identified Pipe is not beyond the knowledge and analytical ability of the average juror.

IV. Governmental Immunity

¶42 Next, the District argues that summary judgment is appropriate, irrespective of whether the public duty doctrine applies or whether the Jenkinses were required to designate an expert, because the decision about when to replace the Water Line Section was a discretionary function for which the District is immune from suit under the GIAU. See Utah Code Ann. § 63G‐7‐301(5)(a) (2011) (providing an exception to the waiver of governmental immunity for injuries arising out of “the exercise or performance, or the failure to exercise or perform, a discretionary function, whether or not the discretion is abused”). While “a party’s entitlement to discretionary function immunity is a question of law,” it is dependent upon the particular facts and circumstances of the challenged act or failure to act by the public entity. See Laney v. Fairview City, 2002 UT 79, ¶ 16, 57 P.3d 1007 (plurality opinion).

A. The Development of Governmental Immunity in Utah

¶43 At common law, a governmental entity was not immune from liability for its negligence in the provision of water. In Egelhoff v. Ogden City, 71 Utah 511, 267 P. 1011 (1928), the Utah Supreme Court affirmed the denial of the city’s motion for a directed verdict on the plaintiff’s claim that the city negligently maintained its water lines, which resulted in leaks that caused damage to the plaintiffs’ property. See id. at 1012‐17. The supreme court explained that

where the water system of a municipal corporation is conducted by the municipality in part for profit, even if principally used for public purposes, the municipality acts in its corporate or private capacity and is liable for damages caused by its negligent construction or management, to its employee or the public

20100400‐CA 22 generally, to the same extent as a private individual or corporation would be under the circumstances. But, there is a distinction between furnishing water to individuals for compensation and furnishing it for fire purposes. The former is the exercise of a private, the latter a governmental, function; and there is no liability if the negligent act was done in the extinguishment of fire, or in connection with flushing hydrants solely to better fire protection, or the like.

Id. at 1012 (emphases added);11 see also 18A McQuillin Municipal Corporations § 53.103‐ 105 (3d ed. 2002) (noting that although a governmental entity is not liable for negligently supplying water for public purposes such as protecting “against fire, flushing sewers, and other uses pertaining to the public health and safety,” it is liable for ordinary negligence arising out of its activity of supplying water for a fee). Thus, the Egelhoff court distinguished between activities engaged in for public purposes and those undertaken for profit, concluding that only the former were entitled to immunity. See Egelhoff, 267 P. at 1012‐13.

¶44 For a period of time after Egelhoff, the Utah Supreme Court continued to distinguish between governmental and proprietary activities in determining whether a governmental entity could be held liable for negligence related to the provision of water. See, e.g., Nestman v. South Davis City Water Improvement Dist., 16 Utah 2d 198, 398 P.2d 203, 205 (1965) (concluding that a water improvement district was not entitled to sovereign immunity for damages caused by the breach of its reservoir, because “operating a water system and supplying water for fees . . . is a proprietary function, and the city is liable for damage or injury caused by its negligence in connection therewith”); Gordon v. Provo City, 15 Utah 2d 287, 391 P.2d 430 (1964) (affirming a jury verdict for a plaintiff injured by a loose water meter lid because the city “operated the water system as a commercial venture in a proprietary capacity and thus [was] responsible for any negligence in operating or maintaining it”). Thus, under Utah’s

11 In Brown v. Salt Lake City, 33 Utah 222, 93 P. 570 (1908), abrogated on other grounds by Kessler v. Mortenson, 2000 UT 95, ¶¶ 13‐14, 16 P.3d 1225, the Utah Supreme Court similarly concluded that the maintenance of a conduit that was part of a waterworks system that the city was not required to furnish to its inhabitants was not “owned, maintained, nor operated in a governmental capacity.” See id. at 574.

20100400‐CA 23 common law, a governmental actor was liable for negligence only with respect to its proprietary activities. See also Standiford v. Salt Lake City, 605 P.2d 1230, 1233 (Utah 1980) (collecting early Utah decisions using the distinction between governmental and proprietary functions to determine whether the governmental actor could be held liable in tort); DeBry v. Noble, 889 P.2d 428, 440 (Utah 1995) (discussing the development of governmental immunity in Utah).

¶45 In 1965, the Utah Legislature adopted the Utah Governmental Immunity Act (the UGIA), which codified the instances in which the traditional immunity of governmental entities is waived. See DeBry, 889 P.2d at 432. The UGIA provided immunity for governmental entities engaged in governmental functions unless expressly waived. See id. Although “the term ‘governmental function’ was the operative phrase establishing the scope of governmental immunity,” the 1965 version of the UGIA did not define that term. See id. Thus, to determine whether a governmental entity was engaged in a governmental function for purposes of the UGIA, the courts borrowed the distinction between proprietary and governmental functions that had originated under the common law. See Greenhalgh v. Payson City, 530 P.2d 799, 801 (Utah 1975) (holding that the UGIA did not displace the common law distinction between proprietary and governmental functions), superseded by statute as stated in Standiford, 605 P.2d at 1238.

¶46 In 1980, the Utah Supreme Court refined the methodology for determining the scope of governmental immunity by considering whether “the activity under consideration is of such a unique nature that it can only be performed by a governmental agency or that it is essential to the core of governmental activity.” See Standiford, 605 P.2d at 1236‐37. During this period, the operation of a municipal water system was considered a proprietary function, falling outside the protection of the UGIA. See Bennett v. Bow Valley Dev. Corp., 797 P.2d 419, 422 (Utah 1990).12 Shortly thereafter, the legislature amended the UGIA to include a statutory definition of governmental function, thereby superceding the Standiford decision. See Utah Code Ann. § 63‐30‐2(4) (Supp. 1987). The 1987 amendment significantly expanded the definition of governmental function by including all governmental acts irrespective of whether they were “essential,” “core,” or “unique to government.” See id. § 63‐30‐

12 Although Bennett v. Bow Valley Development Corp., 797 P.2d 419 (Utah 1990), was decided after the 1987 amendments to the UGIA, it arose prior to those amendments and was decided based on the pre‐1987 precedent. See id. at 421.

20100400‐CA 24 2(4)(a). Subsequent amendments and the reenactment of the UGIA as the GIAU have reiterated the legislature’s all‐inclusive definition of a governmental function for purposes of governmental immunity. See id. § 63G‐7‐102(4) (2011). With this historical framework in mind, we proceed to our analysis of whether the district is immune from suit.

B. The Little Test

¶47 To determine if a governmental entity or its employees are entitled to statutory immunity, the courts undertake a three‐part analysis. See Ledfors v. Emery Cnty. Sch. Dist., 849 P.2d 1162, 1164 (Utah 1993); Clegg v. Wasatch Cnty., 2010 UT 5, ¶ 11, 227 P.3d 1243. First, we must determine whether the “activity . . . performed by [the entity is] a governmental function and therefore immunized from suit by the general grant of immunity contained in [section 63G‐7‐201 of the GIAU].” Clegg, 2010 UT 5, ¶ 11; see also Utah Code Ann. § 63G‐7‐201. Next, if the activity is a governmental function, we must determine whether “some other section of the [GIAU] has waived that blanket immunity.” Ledfors, 849 P.2d at 1164. Finally, “if the blanket immunity has been waived,” we then must determine whether the GIAU contains “an exception to that waiver which results in a retention of immunity against the particular claim asserted in this case.” Id.

¶48 Under the GIAU’s initial grant of immunity, governmental entities and their employees “are immune from suit for any injury that results from the exercise of a governmental function.” See Utah Code Ann. § 63G‐7‐201. The legislature has defined “[g]overnmental function” to mean “each activity, undertaking, or operation of a governmental entity,” including those “performed by a department, agency, employee, agent, or officer of a governmental entity,” and also “includes a governmental entity’s failure to act.” Id. § 63G‐7‐102(4)(a)‐(c). The parties here do not dispute that the District is a governmental entity or that the maintenance of the water line is a governmental function under the broad definition contained in the GIAU. For purposes of its summary judgment argument, the District also concedes that immunity would be waived under one of two waiver provisions of the GIAU. See id. § 63G‐7‐301(3)(a)(i) (providing that immunity is waived for any injury caused by “a defective, unsafe, or dangerous condition of any highway, road, street, alley, crosswalk, sidewalk, culvert, tunnel, bridge, viaduct, or other structure located on them”); id. § 63G‐7‐301(a)(ii) (providing that immunity is waived for injury caused by “any defective or dangerous condition of a . . . public improvement”). The point of contention between the parties is

20100400‐CA 25 the District’s assertion that its conduct falls “under the [GIAU’s] discretionary function and latent defective condition exceptions to the GIAU’s waiver of immunity.”

¶49 The GIAU provides that immunity is not waived, and instead retained, despite any waiver provision in the statute, “if the injury arises out of, in connection with, or results from . . . the exercise or performance, or the failure to exercise or perform, a discretionary function, whether or not the discretion is abused.” Id. § 63G‐7‐301(5)(a). “Discretionary function immunity is a distinct and limited form of immunity that should be applied only when a plaintiff is challenging a governmental decision that involves a basic policy‐making function.” Sandberg v. Lehman, Jensen & Donahue, LC, 2003 UT App 272, ¶ 10, 76 P.3d 699 (internal quotation marks omitted). “[D]iscretionary function immunity is intended to shield [only] those governmental acts and decisions impacting on large numbers of people in a myriad of unforeseeable ways from individual and class legal actions, the continual threat of which would make public administration all but impossible.” Id. (internal quotation marks omitted); see also Hansen v. Salt Lake Cnty., 794 P.2d 838, 846 (Utah 1990). However, because “[n]early all acts performed by government employees involve some amount of discretion,” Nelson, 919 P.2d at 575, courts “read the discretionary function exception to the immunity waiver narrowly,” Johnson v. Utah Dep’t of Transp., 2006 UT 15, ¶¶ 19, 21, 133 P.3d 402. Furthermore, “[i]mmunity is an affirmative defense which the defendant bears the burden of proving.” Trujillo v. Utah Dep’t of Transp., 1999 UT App 227, ¶ 27, 986 P.2d 752.

¶50 The Utah Supreme Court has identified a four‐part test (the Little test or the Little factors) to decide whether the discretionary function exception to the immunity waiver applies:

(1) Does the challenged act, omission, or decision necessarily involve a basic governmental policy program or objective?

(2) Is the questioned act, omission, or decision essential to the realization or accomplishment of that policy, program, or objective as opposed to one which would not change the course or direction of the policy, program, or objective?

20100400‐CA 26 (3) Does the act, omission, or decision require the exercise of basic policy evaluation, judgment, and expertise on the part of the governmental agency involved?

(4) Does the governmental agency involved possess the requisite constitutional, statutory, or lawful authority and duty to do or make the challenged act, omission, or decision?

Johnson, 2006 UT 15, ¶ 22 (quoting Little v. Utah State Div. of Family Servs., 667 P.2d 49, 51 (Utah 1983)). The determination of whether an action of a governmental entity involves a discretionary function is a factually intensive inquiry and typically is not amenable to resolution by summary judgment. See Trujillo, 1999 UT App 227, ¶ 28 (citing Hansen, 794 P.2d at 846); see also Laney v. Fairfield City, 2002 UT 79, ¶ 16, 57 P.3d 1007 (“Although a party’s entitlement to discretionary function immunity is a question of law, a court must have sufficient facts before it to determine whether the challenged act, omission, or decision satisfies the four‐part Little test.”).

C. There Are No Disputed Questions of Fact Regarding Whether the District’s Decision About When to Replace the Water Line Section Was a Discretionary Function

¶51 Nevertheless, the District contends that summary judgment is appropriate here because the undisputed facts demonstrate that the decision of when to replace the Water Line Section was a discretionary function as a matter of law. Because the District bears the burden of proof at trial on the issue of immunity, see Trujillo, 1999 UT App 227, ¶ 27, it had the initial burden “to provide the court with facts that demonstrate both that [it] is entitled to judgment as a matter of law and that there are no material issues of fact that would require resolution at trial,” see Orvis v. Johnson, 2008 UT 2, ¶ 19, 177 P.3d 600. In support of its motion for summary judgment, the District provided a detailed affidavit from Alan Packard, the District’s Assistant General Manager and Chief Engineer,13 setting forth the analysis employed by the Engineers to identify

13 Packard was previously employed as the Engineering Department Manager for the District. It was in his capacity as Engineering Department Manager that he evaluated the Water Line Section.

20100400‐CA 27 pipeline sections in need of replacement. Packard also outlined the factors the Board weighed to determine the order in which the various Identified Pipes were actually replaced, including the Board’s allocation of a finite pool of annual resources for that purpose and its attempts to coordinate with municipal road projects so as to minimize the cost and inconvenience of replacement. Packard further described the way in which these procedures affected both the date when the Water Line Section was first recommended for replacement and the date it was actually replaced.

¶52 Once the District set forth the facts supporting its discretionary function defense, the Jenkinses were required to “set forth specific facts showing that there is a genuine issue for trial.” See id. ¶ 18 (internal quotation marks omitted). Notwithstanding that obligation, the Jenkinses have made no attempt to offer any evidence that calls into dispute the affidavit testimony provided by the District concerning the methodology employed to prioritize the replacement of the Water Line Section.14 Instead, the Jenkinses challenged two of the facts set forth by the District in its motion for summary judgment on the grounds that they were “speculative, without foundation and not relevant.”15

¶53 First, the Jenkinses objected to the District’s statement in its motion for summary judgment that “[b]ecause it is financially impossible for [the District] to replace all of its pipelines in its water system that are experiencing breaks or approaching their useful lifetime, the determination of which aging pipeline within its water system will be replaced is the subject of intensive annual scrutiny by [the District].” With respect to this fact, the Jenkinses argued that

14 Consequently, the Jenkinses also failed to comply with rule 7(c)(3)(B) of the Utah Rules of Civil Procedure, which requires that for each fact the nonmoving party claims is disputed, that party “shall provide an explanation of the grounds for any dispute, supported by citation to the relevant materials, such as affidavits or discovery materials.” See id. 15 None of these objections is well taken. Packard adequately describes his personal involvement in the budget process, his familiarity with the volume of cast iron pipe in the water line, and the factors considered in determining the order in which sections of the pipeline are replaced.

20100400‐CA 28 [f]rom the remaining [f]acts listed by [the District], it is obvious that the focus of the meetings held to discuss which pipelines to replace, is on the financial aspects of the job, not the safety or the reliability of the system. The “scrutiny” described in . . . Packard’s [a]ffidavit does not appear to be “intense” at all, but just a review of the proposed budget and “when can the repairs be done the most cheaply.”

Rather than disputing the fact that the District engaged in the analysis described by Packard, the Jenkinses argue that the District’s analysis was inadequate. This argument does not create a dispute as to the facts asserted in the affidavit. See Chapman v. Primary Children’s Hosp., 784 P.2d 1181, 1186 (Utah 1989) (“[M]ere conclusory allegations in a pleading, unsupported by a recitation of relevant surrounding facts, are insufficient to preclude dismissal or summary judgment.”). Rather, it may have some relevance to the legal questions raised by this dispute, including whether the District was engaged in a discretionary function for purposes of the GIAU.

¶54 Second, the Jenkinses challenged the District’s statement that “it is financially impossible for [the District] to replace every pipeline in its water system that has experienced a break or is approaching its useful lifetime.” As support, the Jenkinses assert that the “[D]istrict is in the business of providing a service to both wholesale and retail customers and presumably has the ability to charge its customers enough for those services to meet its reasonable and necessary expenditures.” However, the Jenkinses again fail to provide contrary evidence to dispute the affidavit testimony. Instead, they speculate that the Board could have raised water fees to a level that would have allowed the District to replace all of the cast iron pipe in the water line immediately. Such unsupported opinion is not sufficient to create an issue of fact. See Treloggan v. Treloggan, 699 P.2d 747, 748 (Utah 1985) (per curiam) (rejecting an affidavit in opposition to a motion for summary judgment because it “reveal[ed] no evidentiary facts, but merely reflect[ed] the affiant’s unsubstantiated opinions and conclusions”).

¶55 In both of its challenges to the District’s statement of facts, the Jenkinses simply attack the wisdom of the Board’s budgetary decisions. See generally Utah Code Ann. § 63G‐7‐301(5)(a) (2011) (providing an exemption to the waiver of immunity for discretionary functions “whether or not the discretion is abused”). Thus, they have not met their burden of coming forward with nonspeculative facts to challenge the District’s evidence. “Where the movant supports a motion for summary judgment with affidavits

20100400‐CA 29 or other sworn evidence, the nonmoving party may not rely on bare allegations from the pleadings to raise a dispute of fact.” Poteet v. White, 2006 UT 63, ¶ 7, 147 P.3d 439. Because the Jenkinses failed to provide any evidence contrary to Packard’s affidavit, we conclude that there are no material issues of fact in dispute. We, therefore, now consider whether the District has met its burden of showing that it is entitled to judgment as a matter of law given these undisputed facts. See Orvis, 2008 UT 2, ¶ 19. To do so, we examine each of the four Little factors to determine whether the decision of when to replace the Water Line Section involved a discretionary function. See Johnson v. Utah Dep’t of Transp., 2006 UT 15, ¶ 22, 133 P.3d 402; Little v. Utah State Div. of Family Servs., 667 P.2d 49, 51 (Utah 1983).

D. Application of the Little Test

¶56 Under the first part of the Little test, we must determine if “the challenged action, omission, or decision necessarily involve[s] a basic governmental policy, program, or objective.” See Johnson, 2006 UT 15, ¶ 22. The District is a water conservancy district formed pursuant to the Water Conservancy District Act. See Utah Code Ann. § 17B‐2a‐ 1001 to ‐1009 (2009 & Supp. 2011).16 Water conservancy districts are created, among other reasons, to conserve and develop water in the state; to provide for its beneficial use; to control and make use of it for domestic, manufacturing, irrigation, and power; to promote the prosperity and general welfare of the people of the state; and to “construct, finance, operate, and maintain works in the state.” See id. § 17B‐2a‐1002(1)(a)‐(f) (2009). The Water Conservancy District Act grants water districts significant authority, including the power to “acquire or construct works, facilities, or improvements.” Id. § 17B‐2a‐1004(1)(c) (Supp. 2011); see also id. § 17B‐1‐102(35) (defining “works” as “a dam, reservoir, well, canal, conduit, pipeline, drain, tunnel, power plant, and any facility,

16 The Water Conservancy District Act, was rearranged and reenacted in 2007. See Special & Local Districts Amendments, ch. 329, §§ 359‐66, 2007 Utah Laws 1723, 1894‐ 1900. However, because the material provisions relevant to this case are substantially the same, we cite to the current version for the convenience of the reader. Compare Utah Code Ann. § 17B‐2a‐1002(1) (2009) (current statute describing the purpose of water conservancy districts), and id. § 17B‐2a‐1004(1)(c) (Supp. 2011) (current statute describing the powers of water conservancy districts), with id. § 17A‐2‐1401 (2004) (prior statute listing the benefits and policies of water conservancy districts), and id. § 17A‐2‐ 1413 (2004) (prior statute defining the powers of water conservancy districts).

20100400‐CA 30 improvement, or property necessary or convenient for supplying or treating water for any beneficial use, and for otherwise accomplishing the purpose of the local district” (emphasis added)). Thus, the District’s decisions concerning the maintenance of the Water Line Section involve a basic governmental policy, program, and objective under the GIAU. Cf. Rocky Mountain Thrift Stores, Inc. v. Salt Lake City Corp., 784 P.2d 459, 463 (Utah 1989) (concluding that decisions regarding “the design, capacity, and construction” of a flood control system involved a basic governmental objective of “flood control to protect life and property”).

¶57 Second, we consider whether the decision to delay replacement of the pipe for three years after the Water Line Section had been recommended for replacement is “essential to the realization or accomplishment of that policy, program, or objective as opposed to one which would not change the course or direction of the policy, program, or objective.” See Johnson, 2006 UT 15, ¶ 22. Here, the policy, program, or objective is the provision of water for beneficial use through 275 miles of pipe comprising the water line. The District contends that the decisions concerning when to replace the numerous sections of Identified Pipe are essential to the accomplishment of its objectives. The Jenkinses disagree, asserting that “[t]he timing of the replacement is not essential to the objective of providing water to [the District’s] customers.” While the District contends that it must prioritize because it cannot replace all segments of its water line that are approaching the end of their useful life at once, the Jenkinses assert that the District could simply charge high enough fees to replace all of the aging pipe.17 However, decisions concerning what amount of an entity’s resources should be allocated to repairs or improvements involve a classic exercise of discretion. See id. ¶ 26 n.10 (acknowledging that a governmental entity “may qualify for the discretionary function exception by demonstrating that a particular allocation of funds was based on an analytical project prioritization”).

¶58 In Duncan v. Union Pacific Railroad, 842 P.2d 832 (Utah 1992), the supreme court held that the Utah Department of Transportation (UDOT) was immune from a

17 Although the parties engage on the issue of whether the District could replace all of the sections of the 275‐mile water line that are comprised of cast iron pipe, the Water Line Section had actually been identified for replacement. Thus, the issue is whether the District could have replaced all of the Identified Pipe, not whether it could have replaced all of the cast iron pipe in the system.

20100400‐CA 31 wrongful death suit brought by the heirs of persons killed in an automobile‐train collision at a rural railroad crossing. See id. at 835‐36. The plaintiffs claimed that UDOT was negligent in deferring the installation of automatic warning lights, signs, and gates at the crossing, and that the decision to do so was an operational, rather than a discretionary, decision. See id. at 833‐34. In rejecting that argument, the Utah Supreme Court described UDOT’s process of installing upgrades. See id. at 834. UDOT first assigned priority according to the hazardousness of the particular crossing. See id. It then installed the upgrades in that order until its federal funding was exhausted, which generally covered “only . . . eight to ten projects in Utah each year.” See id. Applying the Little test, the Duncan court concluded that “the evaluation of crossings and the assigning of priorities for upgrading the adequacy of warning devices . . . [was] essential to the improvement of public safety.”18 Id. at 835.

¶59 This court reached a similar conclusion in Smith v. Weber County School District, 877 P.2d 1276 (Utah Ct. App. 1994), where we determined that a school district was entitled to discretionary function immunity for its decision to locate a school bus stop on a particular side of a busy road. See id. at 1281‐82. In conducting the Little test, the Smith court emphasized that “the bus stop and route in question were part of a vast system serving an entire school district” and that the design of the system involved competing concerns such as “safety, feasibility, and cost effectiveness.” See id. at 1281. Consequently, the court concluded that the system itself was essential to accomplishing the goals of transporting children to school and educating them. See id. at 1281‐82.

¶60 Viewing the Water Line Section as part of the District’s entire water distribution system, see id. at 1281, we conclude that the assignment of priorities for replacement of Identified Pipe is essential to the District’s objectives. See Duncan, 842 P.2d at 835. The Jenkinses’ argument that the District could raise its fees and replace all of the pipe immediately ignores the other factors the District weighs in deciding when a particular section of Identified Pipe will be replaced. Cf. Laney v. Fairview City, 2002 UT 79, ¶¶ 10, 18, 67, 57 P.3d 1007 (plurality opinion) (noting that a municipal power system had the

18 The supreme court also concluded that UDOT’s activities met the other Little test factors, concluding that UDOT had exercised “basic policy evaluation, judgment and expertise,” and that it had the authority “to determine which crossings [were] most hazardous and most deserving of the limited funds available for active warning devices.” See Duncan v. Union Pac. R.R., 842 P.2d 832, 835 (Utah 1992).

20100400‐CA 32 ability to raise fees, but not considering that fact in determining that its decisions about the features of power lines, including the minimum height of high voltage power lines, were essential to the governmental objective of promoting public safety). For example, the District’s Assistant General Manager/Chief Engineer explained in his affidavit that the District may be unable to obtain the necessary permits from the relevant municipality if there has already been recent construction in the area. He further stated that the District seeks to avoid “unnecessary demolishment of recent improvements to property, roads, sidewalks, and landscaping,” and the disturbance to the public caused by “ongoing and repeated construction in the same area.” Indeed, when the Water Line Section was replaced, it was done in conjunction with a road project scheduled for the same section of 3300 South. “‘The decision of how and when to replace a major element of a substantial public facility is . . . at bottom a question of how best to allocate resources.’” Keegan v. State, 896 P.2d 618, 625 (Utah 1995) (quoting Baum v. United States, 986 F.2d 716, 724 (4th Cir. 1993)). Thus, we conclude that the second part of the Little test is met because the prioritization of replacement projects was essential to the operation of the water system.

¶61 Under the third prong of the Little test, the District must show that decisions concerning when to replace the Water Line Section “require[ed] the exercise of basic policy evaluation, judgment, and expertise on the part of the governmental agency involved.” Little v. Utah State Div. of Family Servs., 667 P.2d 49, 51 (Utah 1983) (internal quotation marks omitted). The parties point us to two decisions from our supreme court that they contend are controlling on the issue, Keegan v. State, 896 P.2d 618 (Utah 1995), and Johnson v. Utah Department of Transportation, 2006 UT 15, 133 P.2d 402. We conclude that under either of these opinions, the District’s decision regarding the priority of replacing the Identified Pipe is a discretionary function.

¶62 In Keegan, the plaintiff brought a wrongful death action, claiming that UDOT was negligent in not raising the concrete barrier separating the eastbound and westbound lanes of Interstate Highway 80. See Keegan, 896 P.2d at 619. Although the barrier met national safety standards when first constructed, subsequent projects had added road surface material, reducing the distance between the top of the barrier and the road. See id. At the time of the accident, in which plaintiff’s husband was killed, the barrier did not meet national safety standards. See id. As a result, the plaintiff claimed that UDOT negligently failed to raise the height of the barrier during the overlay projects. See id. The trial court denied UDOT’s claim that its decision was a discretionary function, and

20100400‐CA 33 the jury entered a verdict in favor of the plaintiff. See id. UDOT appealed, claiming, among other issues, that it was entitled to discretionary function immunity. See id.

¶63 Applying the Little factors, the Utah Supreme Court first concluded that UDOT’s decision involved the basic governmental objective of public safety on the roads. See id. at 624. It then determined that the decision was essential to that policy because “it involved a determination of not only the degree of safety that would be provided by various options considered, but also what degree of safety would be an appropriate goal given the time and cost constraints.” Id. Where the project design engineer had prepared a cost‐benefits report based on an extensive safety study examining cost, other scheduled projects, and inconvenience, and where the timing and plans for the project were debated by senior engineers, the supreme court concluded that the decision involved UDOT’s basic policy judgment and expertise. See id. Thus, the Keegan court held that the decision not to raise the barrier was a discretionary function for which UDOT was immune. See id. at 624‐25. The supreme court explained that “‘[e]very highway could probably be made safer by further expenditures, but we will not hold UDOT (and implicitly, the legislature) negligent for having to strike a difficult balance between the need for greater safety and the burden of funding improvements.’” See id. at 624 (quoting Duncan v. Union Pac. R.R., 790 P.2d 595, 601 (Utah Ct. App. 1990), aff’d, 842 P.2d 832 (Utah 1992)). Indeed, the court cited favorably to federal authority on this point, stating,

“The decision of how and when to replace a major element of a substantial public facility is, like the decisions involving design and construction, at bottom a question of how best to allocate resources. Such a decision is inherently bound up in considerations of economic and political policy, and accordingly is precisely the type of governmental decision that Congress intended to insulate from judicial second guessing through tort actions for damages.”

Id. at 625 (emphasis added) (quoting Baum, 986 F.2d at 724); see also Baum, 986 F.2d at 723 (holding that a decision not to replace a guardrail on a National Service Parkway was a discretionary function).

¶64 The Jenkinses claim that Keegan can be distinguished because the decision to wait three years before replacing the Water Line Section did not involve the same type of

20100400‐CA 34 careful analysis. Instead, they point to the supreme court’s decision in Johnson as authority for the proposition that the District’s decision was merely operational. See Johnson, 2006 UT 15, ¶¶ 35‐37. There, the plaintiff was injured when he drove into cutouts running parallel to his lane of travel on Interstate Highway 15 (I‐15) that were marked only by orange barrels. See id. ¶ 4. UDOT’s Regional Director decided to use the plastic barrels without requesting any safety analysis and without discussing it with his supervisor. See id. ¶ 5. Furthermore, he did so despite the contractor’s request to erect concrete barriers instead, and the concern of UDOT’s Project Engineer that the plastic barrels were not adequate for safety purposes. See id. ¶¶ 6‐7. Indeed, the contractor had requested additional funds to cover the cost of switching to concrete barriers, but no agreement on price was ever reached.19 See id. ¶¶ 7‐8. When the plaintiff sued for the substantial injuries he suffered when his car fell into the cutout, the trial court granted summary judgment in favor of UDOT on the basis of the discretionary function exception to the waiver of governmental immunity. See id. ¶ 12.

¶65 On appeal, the supreme court reversed, explaining that a governmental entity cannot escape liability by merely asserting that some minimal level of discretion was used in making a decision. See id. ¶ 21. Rather, “[t]he key . . . is that the government actually exercises a level of discretion in a manner that implicates policy‐making and thrusts the decision into the political process.” Id. (citing Keegan, 896 P.2d at 623‐24); see also Little, 667 P.2d at 51 (“If the State posits immunity on . . . an exercise of discretion, it must make a showing that a conscious balancing of risks and advantages took place.”). The court explained that “[a] real and substantial consideration of the relevant concerns, such as the benefits to public welfare and safety weighed against increases in spending taxpayers’ money, is necessary to prove that the government engaged in policy‐level considerations.” Johnson, 2006 UT 15, ¶ 36. To illustrate the application of that test, the Johnson court distinguished between the initial decision to undertake the construction project and the later decision to use plastic barrels to mark hazards during that project.

19 During the project, another car drove into the cutouts and knocked over a number of orange barrels. See Johnson v. Utah Dep’t of Transp., 2006 UT 15, ¶ 10, 133 P.2d 402. But for the fact that the contractor’s employees had temporarily left the site, they likely would have been struck by the automobile. See id. After that event, the contractor again questioned the adequacy of the existing safety measures. See id. However, the Regional Director made no changes prior to the plaintiff’s accident. See id. ¶¶ 10‐11.

20100400‐CA 35 See id. ¶¶ 34‐35. With respect to the decision “whether or not to perform the construction on I‐15,” the supreme court concluded that “the decision to spend nearly five million dollars on the construction project . . . combined with the adverse effects on public convenience, placed that decision squarely in the public policy arena.” Id. ¶ 34. In contrast, the placement of orange barrels “was . . . so inconsequential in the overall construction project” that UDOT supervisors were not even consulted before the barrels were used. See id. The supreme court noted “[t]he manner in which the construction occurred . . . is where the line is drawn between operational and policy decisions.” Id. ¶ 35; see also Trujillo v. Utah Dep’t of Transp., 1999 UT App 227, ¶¶ 23‐27, 30‐34, 986 P.2d 752 (collecting cases holding that because UDOT had not established that its decisions regarding a construction project were made at the policy level, it was not entitled to rely on the discretionary function exception to the waiver of immunity under the UGIA). Because the decision to use the plastic barrels was an operational decision, the supreme court held that UDOT could not rely on the discretionary function exception to the waiver of immunity. See Johnson, 2006 UT 15, ¶¶ 37, 39 (concluding that the “unsubstantiated decision” of the regional director “not to spend money [was] not the same as a thorough and deliberate cost‐benefit analysis”). Like UDOT in Johnson, the District’s decision “to spend nearly [$180,000] on the construction project, . . . combined with the adverse effects on public convenience, placed that decision squarely in the public policy arena.” See id. ¶ 34.

¶66 The District has provided the only evidence of the criteria used and the process followed to determine the order in which the sections of Identified Pipe are replaced. After gathering data on the pipelines in the District’s system, the Engineers meet and determine which sections need to be replaced, which sections they will recommend to the Board for immediate replacement, and which sections they will recommend to be deferred for replacement at a later date. In doing so, they consider cost, budgetary considerations, permitting issues, soil conditions likely to cause greater degradation of the pipe, the history of breaks, and the possibility of coordinating with other construction projects so as to conserve resources and minimize disruption. The Engineering Department Manager then presents these recommendations to the Board, including the supporting rationale. The Board decides which replacement projects it will fund.

¶67 Based on this evidence, the District has established that the replacement of Identified Pipe “was based on an analytical project prioritization.” Johnson v. Utah Dep’t of Transp., 2006 UT 15, ¶ 26 n.10, 133 P.2d 402. The Jenkinses identify no contrary

20100400‐CA 36 evidence; instead they assert that the failure to consider the risks to third parties precludes a conclusion that the District exercised discretion at a policy‐making level. However, the allocation of limited resources among capital improvement projects “‘is precisely the type of governmental decision that [the legislature] intended to insulate from judicial second guessing through tort actions for damages.’” See Keegan v. State, 896 P.2d 618, 625 (Utah 1995) (quoting Baum v. United States, 986 F.2d 716, 724 (4th Cir. 1993)). Consequently, we conclude that the decision of when to replace the various sections of Identified Pipe, including the Water Line Section, involved the basic policy judgment and expertise of the District.

¶68 Under the fourth Little factor, we ask, “Does the governmental agency involved possess the requisite constitutional, statutory, or lawful authority and duty to do or make the challenged act, omission, or decision?” See Little v. Utah Div. of Family Servs., 667 P.2d 49, 51 (Utah 1983). Utah Code section 17B‐1‐103(2)(d) provides that a local water district “may acquire or construct works, facilities, and improvements necessary or convenient to the full exercise of the district’s powers, and operate, control, maintain, and use those works, facilities, and improvements.” Utah Code Ann. § 17B‐1‐103(2)(d) (Supp. 2011). Therefore, the District has the express authority to replace sections of the water line.

¶69 The undisputed facts indicate that the District’s decisions concerning the order in which Identified Pipe is replaced meets all four of the Little factors. Therefore, we hold that the District is immune from liability to the Jenkinses under the GIAU.20

V. Open Courts

20 The District also asserts that it has retained immunity under the GIAU because the condition of the water line was a latent defect. A latent defect is “a defect which reasonably careful inspection will not reveal.” Pigs Gun Club, Inc. v. Sanpete Cnty., 2002 UT 17, ¶ 26, 42 P.3d 379 (internal quotation marks omitted). Here, the Jenkinses’ claim is not that the District failed to identify the Water Line Section as needing to be replaced. According to the Jenkinses, the District correctly identified the pipe as defective, but then negligently waited three years before replacing it. Thus, the latent defect exception from the waiver of immunity is not implicated and we do not address it further.

20100400‐CA 37 ¶70 The Jenkinses contend that if the District is immune from liability under the discretionary function exception of the GIAU, the statute violates article 1, section 11 of the Utah Constitution (the open courts clause), which states,

All courts shall be open, and every person, for an injury done to him in his person, property or reputation, shall have remedy by due course of law, which shall be administered without denial or unnecessary delay; and no person shall be barred from prosecuting or defending before any tribunal in this State, by himself or counsel, any civil cause to which he is a party.

Utah Const. art. 1, § 11. Under the open courts clause, the “citizens of Utah have a right to a remedy for an injury,” Judd v. Drezga, 2004 UT 91, ¶ 10, 103 P.3d 135, and “the legislature [is] not . . . free to arbitrarily eliminate common law rights without establishing significant social and policy need or providing reasonable alternatives for the protection and vindication of those rights,” Laney v. Fairview City, 2002 UT 79, ¶ 48, 57 P.3d 1007 (plurality opinion).

A. Abrogation of a Remedy

¶71 In Berry v. Beech Aircraft Corp., 717 P.2d 670 (Utah 1985), the supreme court outlined a test to determine whether a particular piece of legislation violates the open courts provision. See id. at 674‐81. Before implementing the Berry test, however, we generally “first determine whether a cause of action has been abrogated by the legislative enactment.” Laney, 2002 UT 79, ¶ 49. “If no remedy was eliminated, there is no need to proceed with the test set forth in Berry.” Id. In making this threshold determination, we look “to the general law of rights and remedies at the time that the [l]egislature [purportedly] abrogate[d] a remedy.” Day v. State, 1999 UT 46, ¶ 38, 980 P.2d 1171.

¶72 The Jenkinses contend that by defining essentially all activities of a governmental entity as governmental functions, the Utah Legislature abrogated their right to sue the District for negligence. According to the Jenkinses, operating a water system would not have been a governmental function prior to the adoption of that definition and therefore, they would have been able to recover their damages. Indeed, until the 1987 amendments, the operation of a municipal water system was not entitled to immunity

20100400‐CA 38 under the UGIA. See Bennett v. Bow Valley Dev. Corp., 797 P.2d 419, 422 (Utah 1990). For purposes of summary judgment, the District does not challenge the Jenkinses’ assertion, stating that “[e]ven assuming that [the Jenkinses enjoyed a common law right to sue], . . . the 2004 GIA[U] . . . comports with the open courts clause.”21 Therefore, we do not further consider the issue of whether a cause of action was abrogated.

B. Reasonable Alternative Remedy

¶73 Because the District does not dispute that the Jenkinses had a right to sue that is no longer available due to the all‐inclusive definition of governmental function, we now proceed with the Berry test. When a remedy has been abrogated, we must first determine “whether the legislature has provided a ‘reasonable alternative remedy by due course of law for vindication of [a plaintiff’s] constitutional interests.’” Laney, 2002 UT 79, ¶ 54 (additional internal quotation marks omitted) (quoting Berry, 717 P.2d at 680). In order to suffice, the substitute benefit “must be substantially equal in value or other benefit to the remedy abrogated in providing essentially comparable substantive protection to one’s person, property, or reputation, although the form of the substitute remedy may be different.” Berry, 717 P.2d at 680. Here, the Jenkinses have no alternative remedy if the District is immune.

¶74 Nevertheless, the District argues that because the adoption of the UGIA in 1965 broadened governmental liability in some instances (allowing the state, counties, and school districts to be sued) in comparison to the common law, see DeBry v. Noble, 889 P.2d 428, 432 (Utah 1995), we should interpret the 2004 GIAU as providing “the general public collectively with reasonable alternative remedies for any remedies the 2004 [GIAU] may have abrogated.” In addition, during the debates related to the repeal of

21 Thus, we do not undertake the analysis of whether the GIAU abrogated a remedy the Jenkinses would have previously enjoyed. Consequently, we need not consider whether the District’s status as a water district might have created a factual question of whether it was of such a unique character that its activities would have been considered governmental prior to the 1987 amendments to the GIAU. Cf. Parks v. Utah Transit Auth., 2002 UT 55, ¶ 14, 53 P.3d 473 (holding that the Utah Transit Authority, as a public transit district, is of such a unique character, with such expansive and diverse operations, that it would have been a governmental function at common law and therefore, the GIAU did not abrogate plaintiffs’ remedy).

20100400‐CA 39 the UGIA and the enactment of the GIAU in 2004, members of the legislature opined that although the GIAU eliminated remedies for some tort victims, it was a reasonable trade‐off because the bill also “eliminate[d] the procedur[al] traps for the unwary.” See Recording of Utah Senate Floor Debates, 55th Leg., Gen. Sess. (Feb. 24, 2004). Specifically, these legislators noted that the GIAU included more clear and concise notice of claim provisions, codified a statute of limitations discovery rule, and included a safe harbor provision for notices of claims. See id. Accordingly, they concluded that while the bill would leave some tort victims without a remedy, it also “makes it much more likely that numerous claimants who previously lost their claims due to technical failures to meet notice requirements will be able to pursue their claims.” See Recording of House Floor Debates, 55th Leg., Gen. Sess. (Mar. 3, 2004).22 The District claims that this trade‐off is justified by language in Masich v. United States Smelting, Refining & Mining Co., 113 Utah 101, 191 P.2d 612 (1948), indicating that under the open courts clause “certain individual rights and remedies can be made to yield to the public good.” See id. at 624.

¶75 However, this part of the Berry test has not been interpreted so broadly by our supreme court. Instead, under the first part of the Berry test, the legislature “must . . . provide a quid pro quo in the form of . . . a substitute remedy for the individual,” and if this is not provided, the legislature must meet the second part of the Berry test. See Sun Valley Water Beds of Utah, Inc. v. Herm Hughes & Son, Inc., 782 P.2d 188, 192 (Utah 1989) (emphasis added). Where “the [l]egislature has provided non‐common law alternative remedies in lieu of common law remedies that it abrogated, such as in the No‐Fault Insurance Act, the Worker’s Compensation Act, or the Occupation Disease Act,” the supreme court has upheld the statute without proceeding to the next step of the Berry test. Day v. State, 1999 UT 46, ¶ 44, 980 P.2d 1171; see also Masich, 191 P.2d at 624‐25 (upholding the Occupation Disease Act). But it has consistently held that there is not an adequate alternative remedy where “all actions of the type asserted” by a plaintiff are

22 An attorney statement submitted to the Senate committee states, While some tort victims may lose a claim under this bill, the trade off is reasonable because even more victims will be able to pursue claims under the revamped Notice of Claim provisions and other provisions of [the GIAU]. The entire group of tort victims is, therefore, better off, which serves the greater good.

20100400‐CA 40 barred, see Day, 1999 UT 46, ¶ 43, where plaintiffs are limited to alternative claims against different defendants, see Sun Valley Water Beds, 782 P.2d at 192, or where plaintiffs’ proven claims are subject to damages caps, see Judd v. Drezga, 2004 UT 91, ¶ 12, 103 P.3d 135.

¶76 Despite that authority, the District contends that because, by the adoption of the UGIA in 1965, the legislature “simultaneously expanded the immunity of cities and special districts, and restricted the immunity of the State of Utah and its instrumentalities, for the first time, allowing suit against the State,” it provided an adequate alternative remedy. (Emphasis omitted.) However, the Utah Supreme Court rejected a similar argument in Condemarin v. University Hospital, 775 P.2d 348 (Utah 1989). There, a plaintiff challenged the 1978 amendments to the UGIA under the open courts clause, because the amendments added government‐owned hospitals as a governmental entity covered by the UGIA, provided immunity for the hospitals’ proprietary activities, and granted complete immunity to hospital employees. See id. at 351. The University Hospital argued that the governmental immunity statute did not violate the open courts clause because its effect was to expand liability beyond that available at common law in some instances. See id. The Condemarin court disagreed, stating, “Although it is generally true that the [UGIA] expanded government liability, that is not the case with respect to proprietary or nongovernmental functions.” Id. Thus, the court concluded that the amendments had abrogated the plaintiff’s existing remedy without providing a reasonable alternative. See id. at 352. Consequently, we are not persuaded that the expansion of government liability resulting from the 1965 adoption of the UGIA negates the fact that the Jenkinses’ remedy was abrogated by the adoption of an all‐inclusive definition of governmental function. Likewise, the fact that the legislature has made changes that may reduce the number of notices of claim deemed ineffective due to technical deficiencies does not change the impact of the amendments on the Jenkinses’ individual rights.

¶77 Under the GIAU, the Jenkinses are left entirely without a remedy the District concedes would have been available to them before the adoption of the 1987 amendment to the UGIA defining “governmental function.”23 Further, the 2004

23 The 1987 amendments of the UGIA define “[g]overnmental function” as any act, failure to act, operation, function, or undertaking of (continued...)

20100400‐CA 41 enactment of the GIAU retained a similarly all‐inclusive definition24 and did nothing to provide a quid pro quo for persons situated like the Jenkinses. See Sun Valley Water Beds, 782 P.2d at 192. Thus, we conclude that the legislature has not provided the Jenkinses a reasonable alternative remedy.

C. The Laney and Judd Decisions

¶78 Pursuant to the Berry test,

if there is no substitute or alternative remedy provided, abrogation of the remedy or cause of action may be justified only if there is a clear social or economic evil to be eliminated and the elimination of an existing legal remedy is not an arbitrary or unreasonable means for achieving the objective.

Berry v. Beech Aircraft Corp., 717 P.2d 670, 680 (Utah 1985). To determine whether the elimination of the Jenkinses’ remedy meets this standard, we must examine the legislative history of the statute to ascertain the rationale for the subject legislation and then consider whether the means for eradicating the social or economic evils selected is

23 (...continued) a governmental entity whether or not the act, failure to act, operation, function, or undertaking is characterized as governmental, proprietary, a core governmental function, unique to government, undertaken in a dual capacity, essential to or not essential to a government or governmental function, or could be performed by private enterprise or private persons. Utah Code Ann. § 63‐30‐2(4)(a) (1989). 24 The 2004 version of the GIAU defines “[g]overnmental function” as “each activity, undertaking, or operation of a governmental entity”; “each activity, undertaking, or operation performed by a department, agency, employee, agent, or officer of a governmental entity”; and “includes a governmental entity’s failure to act.” Utah Code Ann. § 63‐30d‐102(4)(a)‐(c) (2004).

20100400‐CA 42 arbitrary or unreasonable. See Laney v. Fairview City, 2002 UT 79, ¶¶ 56‐58, 57 P.3d 1007 (plurality opinion) (collecting Utah cases applying this standard). Before we begin that analysis, we pause to examine the appropriate level of deference to the Utah Legislature.

¶79 Of particular interest in undertaking this review is the Utah Supreme Court’s decision in Laney v. Fairview City, 2002 UT 79, 57 P.3d 1007 (plurality opinion). There, a majority of the court held that the 1987 definition of governmental function, see supra note 23, as applied to a municipal power system, violated the open courts clause of the Utah Constitution.25 See generally id. ¶¶ 1, 83. In reaching the conclusion that the statute was unconstitutional, the lead opinion in Laney cited favorably a two‐justice concurrence in Lee v. Gaufin, 867 P.2d 572, 590‐92 (Utah 1993) (Zimmerman, J., and Hall, C.J., concurring in the result),26 in which the concurring justices determined that a presumption of unconstitutionality attached to legislation challenged under the open courts clause. See Laney, 2002 UT 79, ¶ 63; see also Lee, 867 P.2d at 591 (Zimmerman, J., and Hall, C.J., concurring in the result) (“[W]hen we have found a statute to limit a right protected by the open courts [clause], we have, de facto, shifted from a presumption that the limiting statute is constitutional to a presumption that the statute is

25 This was a plurality opinion where Chief Justice Durham wrote the main opinion, joined by Justice Howe. Justice Russon concurred in the result, stating that although the main opinion seemed to accept that the legislature could change the definition of governmental function, governmental functions were inherently different than proprietary functions and the legislature could not similarly “declare automobiles to be sailboats or houses.” See Laney v. Fairview City, 2002 UT 79, ¶ 74, 57 P.3d 1007 (Russon, J., concurring). However, Justice Russon did accept that the legislature could grant governmental immunity for a governmental entity engaged in a proprietary activity by meeting the Berry test. See id. ¶ 79. 26 Although the majority reached its conclusion that the statutes of limitations and repose in the Utah Health Care Malpractice Act were unconstitutional under the uniform operation of laws provision found in article I, section 24 of the Utah Constitution, Justices Zimmerman and Hall determined that the statute of repose’s effect on the claims of minors violated article I, section 11, the open courts clause. See Lee v. Gaufin, 867 P.2d 572, 590 (Utah 1993) (Zimmerman, J., and Hall, C.J., concurring in the result).

20100400‐CA 43 unconstitutional . . . .” (emphasis omitted)); Wood v. University of Utah Med. Ctr., 2002 UT 134, ¶¶ 46‐47, 67 P.3d 436 (Durham, C.J., dissenting) (explaining that “[c]ontrary to the position taken by the lead opinion, this court has consistently rejected the presumption of constitutionality of statutes challenged under the [open courts clause]” in favor of heightened scrutiny). Thus, the Laney court began its analysis with a presumption that the all‐inclusive definition of “governmental function” ran afoul of the open courts clause, and then considered whether that presumption had been rebutted by the municipality. See Laney, 2002 UT 79, ¶ 53.

¶80 In analyzing whether the UGIA violated the open courts clause, the Laney court first noted that the operation of a power system is a dangerous activity and involves “the highest degree of care” to prevent harm to others. See id. ¶ 64 (internal quotation marks omitted). With that backdrop in mind, the court questioned the stated legislative purpose of the amendment, which was “to make liability insurance more affordable for government entities by reducing liability risks.” See id. ¶ 66. The court emphasized that although this may be a worthy objective, no specifics were given about whether a municipal power system in Utah had ever sustained a large damage award or whether their operations had been affected by potential liability. See id. The lead opinion further emphasized that the power system was not subsidized by taxpayers and actually generated an annual profit, concluding that “[i]f the City cannot afford to purchase reasonable amounts of liability insurance to meet its high standard of care, rate increases may be justified and necessary.” See id. ¶ 67. Ultimately, a majority of the court determined that the legislation swept too broadly when it defined “all activities of municipalities as governmental action, regardless of their nature.”27 See id. ¶ 69. The court suggested that the legislature could “create very limited immunities to address

27 Two justices dissented from the open courts analysis in Laney, arguing that the provision should not be interpreted to impose “a strong substantive limitation on the legislature’s ability to limit or eliminate a cause of action for, or the remedies available for personal injury.” Id. ¶ 88 (Wilkins, J., and Durrant, J., dissenting) (internal quotation marks omitted). Instead, the dissenting justices called for the abandonment of the test announced in Berry, due to its intrusion on the legislative function and violation of the separation of powers clause of the Utah Constitution. See id. ¶¶ 88‐106. Finally, because the 1987 amendments to the UGIA did not “unreasonably burden[] important constitutional rights,” the dissenting justices would have upheld it as constitutional. See id. ¶ 138.

20100400‐CA 44 specific problems, or to place ‘caps’ on the amount of damages,” as was upheld in McCorvey v. Utah Department of Transportation, 868 P.2d 41 (Utah 1993). See Laney, 2002 UT 79, ¶ 70; see also McCorvey, 868 P.2d at 47‐48 (holding that the damage cap on claims against governmental entities did not violate the open courts clause because in the absence of the UGIA, the state would have been completely immune).

¶81 Two years later, a majority of the supreme court retreated from the analytical framework applied by the Laney majority, instead adopting a more deferential approach. In Judd v. Drezga, 2004 UT 91, 103 P.3d 135, the court examined a statutory damages cap on quality of life damages in medical malpractice actions adopted with the stated purpose of reducing health care costs. See id. ¶¶ 3, 6. The majority rejected Laney’s presumption of unconstitutionality and instead “recognize[d] an obligation of deference to legislative judgments in a Berry review, and [noted that] to the extent [it] differ[ed] from [the court’s] prior application of Berry, those prior applications [were] disavowed.” See id. ¶ 11. In addition, the Judd court departed from the Laney court’s reassessment of the legislators’ collective judgment that the legislation was necessary to address a legitimate public concern. See id. The higher level of deference afforded the legislature is apparent in the Judd majority’s refusal to second guess the empirical truth of the stated policy considerations by “undertak[ing] the same investigation as the legislature, [and] reviewing its data‐gathering methods and conclusions to determine whether the stated legislative findings are perfectly correct.” See id. ¶ 13. Instead, the Judd majority indicated that “[w]hen an issue is fairly debatable, we cannot say that the legislature overstepped its constitutional bounds when it determined that there was a crisis needing a remedy.” See id. ¶ 15.

¶82 Despite the Judd court’s greater deference to legislative judgments concerning the existence of a crisis needing to be addressed, the majority reaffirmed the part of the Berry test requiring the reviewing court to evaluate whether the means selected for abating that crisis are nonarbitrary and reasonable. See id. ¶ 16; see also Berry v. Beech Aircraft Corp., 717 P.2d 670, 680 (Utah 1985). To do so, the court focused on whether the legislation was “narrowly tailored” to respond to the stated social or economic evils to be abated. See Judd, 2004 UT 91, ¶ 17; Laney, 2002 UT 79, ¶ 57 (quoting Hirpa v. IHC Hosps., Inc., 948 P.2d 785, 793‐94 (Utah 1997) (upholding the Good Samaritan Act and noting that “[t]he statute does not cut an unnecessarily wide swath through causes of action against medical providers”)). Because the damage cap at issue in Judd was narrowly tailored to apply only to quality of life damages, which are “less susceptible to quantification than purely economic damages,” and it could facilitate the “predict[ion]

20100400‐CA 45 and control[ of] future costs [and thereby] result in lower insurance rates,” see Judd, 2004 UT 91, ¶ 17, the supreme court concluded that the damage cap did not violate the open courts clause.28 See id. ¶¶ 39‐40.

¶83 Thus, while the supreme court’s decision in Judd disavows the Laney court’s reassessment of the social or economic evils identified by the legislature, see id. ¶ 15, it embraces the requirement that the legislative remedy adopted be narrowly tailored. See id. ¶ 17; see also Laney, 2002 UT 79, ¶¶ 68‐69. Furthermore, Judd left intact the Laney court’s application of the Berry test to the all‐inclusive definition of “governmental function” adopted in the 1987 amendments to the UGIA, which the court determined “swept too broadly,” see Laney, 2002 UT 79, ¶ 69, despite the fact that governmental entities are subject to liability in many instances only by the grace of the Utah Legislature. See generally Judd, 2004 UT 91, ¶¶ 39‐40; see also Condemarin v. University Hosp., 775 P.2d 348, 351 (Utah 1989) (holding amendments to UGIA unconstitutional under the open courts clause despite expansion of government liability from common law).

¶84 With this precedent in mind, we turn now to the facts at issue here and continue our application of the Berry test. We begin with the identification of the social or economic evils to be eliminated, deferring to the legislature’s judgment and its conclusions concerning the existence of a crisis needing a remedy. See Judd, 2004 UT 91, ¶ 11. We then consider whether the means selected to address that crisis are a reasonable and nonarbitrary method of addressing the elimination of the evils identified. Although we conclude that the means selected are reasonably related to achieving an abatement of the social and economic evils identified by our legislature, we determine that those means, the complete elimination of any remedy, are not narrowly tailored. Therefore, we hold that the abrogation of the Jenkinses’ remedy against the District violates the provisions of the open courts clause of the Utah Constitution.

28 Two justices dissented in Judd v. Drezga, 2004 UT 91, 103 P.3d 135. In particular, they disagreed with the majority’s “deference to legislative judgments in a Berry review” and its rejection of “heightened scrutiny in favor of a rational basis standard” of review. See id. ¶ 42 (Durham, C.J., and Nehring, J., dissenting) (internal quotation marks omitted).

20100400‐CA 46 D. The Social and Economic Evils Identified by the Utah Legislature

¶85 In Laney, the supreme court identified the goals of the 1987 amendments to the UGIA as the dramatic increase in lawsuits naming governmental entities as defendants, large damage awards against governmental entities, and the increasing difficulty for governmental entities to obtain affordable liability insurance. See Laney v. Fairview City, 2002 UT 79, ¶ 65, 57 P.3d 1007. After the Laney court held the 1987 amendments unconstitutional as applied to a municipal power company, the legislature repealed the UGIA and enacted the 2004 GIAU. This was done in direct response to the Laney decision. See Recording of Utah Senate Floor Debates, 55th Leg., Gen. Sess. (Feb. 24, 2004). In enacting the GIAU, the legislature pointed to several social and economic evils it wished to remedy.

¶86 First, the legislature indicated that the “court‐created distinction” between which governmental activities are constitutionally entitled to immunity is “unpredictable and often illogical.” See Recording of House Floor Debates, 55th Leg., Gen. Sess. (Mar. 3, 2004). An Assistant Attorney General, who testified before the Senate Government Operations and Political Subdivisions Committee, stated that tests to distinguish which governmental functions are entitled to immunity “can become traps for the unwary. Both citizens and government can find themselves guessing wrong as to how the courts will draw the line,” and that “[t]he confusion not only has damaged the citizens but it has also injured government.” See id.

¶87 In particular, the legislature was convinced that the historical treatment of the provision of water as proprietary was outdated. During the 2004 debate, the bill’s Senate sponsor stated, “If you went back prior [to Laney] or probably asked any member on this floor, ‘Is a water system a proper role for local government?’—the answer would have been, ‘Well of course, what are you even asking?’” See Recording of Utah Senate Floor Debates, 55th Leg., Gen. Sess. (Feb. 24, 2004).29 Another unidentified senator stated, “I think everyone agrees that nowadays water service, electricity, [and] a lot of utility services are things that government should provide.” Id. And Senator Greg Bell stated, “every water system could be duplicated by a private group or at least a quasi‐

29 The senator was apparently under the impression that prior to Laney, water systems were considered a governmental function and that the decision in that case “threw that all open.”

20100400‐CA 47 private group . . . but we have left municipal government and assigned municipal government those responsibilities.” Id. Furthermore, a document submitted to the Senate committee entitled “Comments on Utah Governmental Claims Act” argues that public utilities, and specifically municipally‐run power systems, have never been proprietary functions.30 It states that

power generation, transmission, and distribution are not and never have been private functions. They are functions of cities and towns, and if cities and towns choose not to perform the functions, they may grant a franchise to a private entity to do so. The Constitution therefore contemplates and makes the provision of public utility service a function and a purpose of government.

¶88 Because the provision of domestic water, unlike water used for fire protection, was traditionally not considered a governmental function for which the entity was immune, we disagree that the Laney analysis created a novel category of liability. See generally Egelhoff v. Ogden City, 71 Utah 511, 267 P. 1011, 1012‐13 (Utah 1928) (holding that the provision of water is a proprietary function, subjecting the city to liability for negligence). However, we acknowledge that much has changed during the eighty‐three years since the Utah Supreme Court decided Egelhoff v. Ogden City, 71 Utah 511, 267 P. 1011 (Utah 1928). While at one time it may have been unusual for a governmental entity to provide water to its inhabitants, by 1990, our supreme court recognized that “[c]ities can and do operate water systems on a commercial basis. In many areas of our state, residents maintain wells and provide their own water. Also, there are privately owned companies supplying water to residents.” Bennett v. Bow Valley Dev. Corp., 797 P.2d 419, 422 (Utah 1990) (citation omitted) (holding “that the maintenance of a water storage tank” was not a governmental function). Thus, we accept the legislature’s judgment that expectations concerning the provision of water may have changed. “[O]ne of the functions of the legislative power is to remedy defects in the common law as they develop, and to adapt to the change of time and circumstance.” Berry v. Beech Aircraft Corp., 717 P.2d 670, 678 (Utah 1985) (internal quotation marks omitted).

30 As previously discussed, our review convinces us that, like most other states, Utah has not historically considered the provision of water for purposes other than fire protection a governmental function. See supra ¶¶ 43‐46.

20100400‐CA 48 However, in summarizing Berry’s holding, the Utah Supreme Court explained that “the rights of individuals protected by the open courts provision must be balanced against the legislature’s need to enact laws to meet changing societal needs.” Day v. State, 1999 UT 46, ¶ 37, 980 P.2d 1171.

¶89 As a second justification for the 2004 GIAU, the legislature indicated its desire to “guard[] government’s ability to act for the public good without fear of unexpected and potentially crippling taxpayer burdens.” Recording of Utah House Floor Debates, 55th Leg., Gen. Sess. (Mar. 3, 2004). An Assistant Attorney General testified, “Government, when it is unsure of whether it has immunity[,] will all too often err on the side of not providing the services that its taxpayers want, but might create a liability that the government cannot afford.”

¶90 Third, the legislature pointed to data that liability insurance premiums have risen for governmental entities dramatically since both the September 11, 2001 attacks and the Utah Supreme Court’s decision in Laney, while the amount of coverage has decreased. See Recording of Utah House Floor Debates, 55th Leg., Gen. Sess. (Mar. 3, 2004) (stating that “since Laney v. Fairview City, the insurance market has responded to court‐injected uncertainty in the governmental immunity realm by increasing premiums 230% for 70% of the prior coverage”).

¶91 In considering the constitutionality of the GIAU as applied to the Jenkinses’ claim, we depart from Laney’s focus on the correctness of the legislature’s identified social and economic evils, instead accepting the supreme court’s more recent recognition in Judd that “our power does not extend so far as to permit imposition of our views on such policy disputes.” See Judd v. Drezga, 2004 UT 91, ¶ 14, 103 P.3d 135; see also id. ¶ 15 (explaining that the courts’ task is not to determine whether the legislature “made wise policy”). Because the social and economic concerns identified by the legislature are “fairly debatable,” we conclude that it has not “overstepped its constitutional bounds” by identifying them as a basis for its abrogation of the Jenkinses’ remedy. See id. ¶ 15. Thus, we accept these social and economic evils as the starting point for our assessment of whether the complete abrogation of the Jenkinses’ remedy is reasonably and rationally related to the elimination of these concerns. Nevertheless, the means adopted to address those identified evils must be narrowly tailored to comply with the open courts clause.

20100400‐CA 49 E. The Reasonableness and Rationality of the Means Selected

¶92 The means adopted by the Utah Legislature to address these identified evils was to designate all action and inaction of any governmental entity as a governmental function for immunity purposes. In doing so, the legislature made it “crystal clear[] that all functions are covered under this governmental immunity” and that the new definition “establishes the legislature’s intent to apply governmental immunity broadly and concisely to all government functions and especially to all discretionary budget decisions that effect the taxing policies and priorities of state and local government.” See Recording of Utah Senate Floor Debates, 55th Leg., Gen. Sess. (Feb. 24, 2004). To determine if the inclusion of all governmental activity within the scope of the GIAU is rationally and reasonably related to the identified concerns, Judd instructs that we must first consider whether “[t]he legislature’s determination that it needed to respond to the perceived . . . crisis was logically followed by action designed” to remedy that crisis. See Judd, 2004 UT 91, ¶ 16. The remedy need not abate the identified evil entirely, but it must be logically designed to remedy the concern in some measurable amount. See id. Furthermore, the means must be narrowly tailored. See id. ¶ 17 (upholding a cap on quality of life damages because “[r]ather than cap all damages, like the cap struck down in Condemarin v. University Hospital, 775 P.2d 348 (Utah 1989), the limitation on recoverable damages in this case is narrowly tailored, by limiting quality of life damages alone”).

¶93 Thus, it is not our role to assess the wisdom of placing the full burden of the District’s decisions to delay the replacement of Identified Pipe on the unfortunate individuals living down‐gradient from the water line, rather than on all customers of the District equally. Cf. Judd, 2004 UT 91, ¶ 17 (“Despite this court’s concerns about the wisdom of depriving a few badly injured plaintiffs of full recovery, the cap is also constitutionally reasonable.”). Moreover, the all‐inclusive definition of governmental function “is not unconstitutionally arbitrary merely because it does so.” See id. ¶ 16. Instead, our review is limited to whether the definition of governmental function, as applied to the District’s activities here, is rationally related, nonarbitrary, and “narrowly tailored” to the stated social evils it was designed to remedy. See id. ¶¶ 16‐17. We explain our reasoning now as to each evil identified by the Utah Legislature.

20100400‐CA 50 1. The Uncertainty About Which Activities Are Immune

¶94 The first concern identified by the legislature is the confusion caused by the unpredictable judicial decisions defining the limits of a governmental function. Although the inclusion of all governmental activity as falling within the GIAU effectively eliminates any confusion on that point, it does not eliminate other factual and legal issues affecting whether governmental activity is immune. As our discussion of discretionary function illustrates, there are other aspects of the GIAU where the law as applied to specific facts will determine whether a particular exception or exemption is triggered. See supra ¶¶ 50‐69. As a result, the all‐inclusive definition of governmental function will not completely prevent uncertainty about whether the entity can be held liable. By including all governmental activity, however, the legislature logically eliminated confusion about whether a particular activity is a governmental function. Therefore, the definition “controls [uncertainty] in one area where [uncertainty] might be controllable.” See id. ¶ 16. In addition, the broad definition includes the provision of water, a function the Utah Legislature mentioned in its debates as being an expected service of municipalities.

¶95 However, the remedy adopted by the legislature must also be narrowly tailored to comply with the open courts clause. See id. ¶ 17. The Utah courts have upheld legislation in the face of an open courts clause challenge when the legislation is carefully constructed so as to leave remedies intact for situations in which the tortfeasor is especially culpable or the plaintiff warrants extra protection. For example, in Craftsman Builder’s Supply, Inc. v. Butler Manufacturing Co., 1999 UT 18, 974 P.2d 1194, the court accepted the legislature’s stated purpose of reducing liability insurance and record storage costs in the construction industry in enacting a builder’s statute of repose, which “provide[d] that actions for injury to persons or property arising out of an improvement to real property must be brought within a certain number of years of the triggering event.” See id. ¶¶ 13, 20. The court then determined that the statute was reasonable and nonarbitrary because the “possibility of injury and damage” after the repose period was “highly remote and unexpected.” See id. ¶¶ 21‐22 (recognizing, with approval, that although data suggested injury was remote after six to ten years, the legislature chose a period of twelve years for the majority of claims). The supreme court further noted that the legislature carved out certain exceptions to the repose period, leaving remedies intact where the provider “fraudulently conceale[d] the act or omission giving rise to the cause of action,” where the “act or omission [was] willful or

20100400‐CA 51 intentional,” where the parties had an express warranty for a longer period of time, or where the plaintiff was a minor or was mentally incompetent. See id. ¶ 22.

¶96 In contrast, the definition of “governmental function” in the GIAU broadly abrogates a large class of previously‐existing actions. See id. ¶ 19 (recognizing that the prior version of the builder’s statute of repose was “‘too likely to cut off injuries that should be compensated’” (quoting Horton v. Goldminer’s Daughter, 785 P.2d 1087, 1095 (Utah 1989)). The legislation at issue in Judd was held to be narrowly tailored because it applied a cap only to a certain type of damages. See Judd v. Drezga, 2004 UT 91, ¶ 17, 103 P.3d 135. By analogy, to achieve the same effect as the definition of governmental function at issue here, the legislation in Judd would have had to bar all damages of any nature and in any amount. It seems unlikely that the supreme court would have upheld the legislation challenged in Judd under those circumstances. See id. ¶ 17; see also Condemarin, 775 P.2d at 349‐52 (holding that the 1978 amendments to the UGIA that added government‐owned health care facilities to the list of governmental entities protected by the statute and made their employees completely immune were unconstitutional under the open courts clause). Thus, the adoption of the all‐inclusive definition of governmental function is not narrowly tailored to address the concern that the court created definition of a governmental function was difficult to predict.

2. The Ability of Governmental Entities to Act for the Public Good Without Incurring Crippling Liability

¶97 The next social evil identified by the Utah Legislature as a motivation for its adoption of the all‐inclusive definition is the chilling effect that the unexpected and potentially crippling liability on the taxpayers will have on a governmental entity’s willingness to act for the public good. Again, because the GIAU waives immunity under other circumstances, the all‐inclusive definition of governmental function is not completely effective in eliminating the fear that actions undertaken for the public good might result in significant liability. Nevertheless, the application of that definition in the context of a water conservancy district does involve the protection of an entity acting for the public good. As previously noted, water conservancy districts are charged with conserving and using water for the benefit of the State. See Utah Code Ann. § 17B‐2a‐1002 (2009). Acknowledging the need for the conservation and beneficial use of this limited resource, the Utah Legislature has adopted a “statutory framework [that] evinces a state policy of displacing competition with regulation in the area of municipal control over water and water rights.” See Salt Lake City Corp. v. Big Ditch

20100400‐CA 52 Irrigation Co., 2011 UT 33, ¶ 64, 258 P.3d 539 (holding that the trial court properly applied the municipal status defense to dismiss antitrust claims against Salt Lake City). Under these circumstances, the inclusion of the District’s activities within the statutory framework of the GIAU rationally promotes the allocation of the State’s water resources for the common good.31

¶98 Notwithstanding that rational connection, the all‐inclusive definition of governmental function is not narrowly tailored to address the specific evil. While the legislature identified the risk of “crippling liability” as the concern to be addressed, the amendment results in the abrogation of even modest damage claims against the District. For example, the District’s annual operating budget for 2006‐2007 was almost $79,000,000, while the Jenkinses’ claim is approximately $116,000. Rather than limiting the amount of recoverable damages, or the categories of damages that can be recovered,

31 Indeed, the supreme court has indicated a willingness to expand the Standiford v. Salt Lake City, 605 P.2d 1230 (Utah 1980), definition of governmental function in the regulation context to account for the qualitative differences between operations for private gain and those undertaken for the public good. See id. at 1233; Madsen v. Borthick, 658 P.2d 627, 630‐31 (Utah 1983) (addressing banking regulation and holding that Standiford “does not preclude governmental immunity for supervisory functions in some respects similar to those that could be performed by a private association authorized by agreement, such as self regulation by an industry”); see also Moss v. Pete Suazo Utah Athletic Comm’n, 2007 UT 99, ¶¶ 23, 26, 175 P.3d 1042 (concluding that governmental regulation of professional boxing “is qualitatively different from the type of regulation that could be provided by a private organization” and that “[o]nly the government can regulate the sport for the common good, rather than for the benefit of a select, financially interested sector of the boxing community”); DeBry v. Noble, 889 P.2d 428, 441 (Utah 1995) (rejecting an argument that private entities could perform building inspections and stating that the argument “ignores the qualitative differences between public and private regulation of building construction”); Gillman v. Department of Fin. Inst., 782 P.2d 506, 512‐13 (Utah 1989) (holding that banking regulation is a uniquely governmental activity and concluding that “governmental entities should not be dissuaded from engaging in [activities undertaken to ensure public health and safety] by the fear that liability may be imposed if an employee performs his duties inadequately” (internal quotation marks omitted)).

20100400‐CA 53 see Judd, 2004 UT 91, ¶ 17, the legislative means adopted eliminates all recovery for any damage caused by the District’s negligent delay in replacing Identified Pipe.

3. The Increasing Number of Lawsuits and Judgments Against Governmental Entities and the Corresponding Increase in Insurance Costs

¶99 The last evil identified as warranting the inclusion of all governmental activities under the umbrella of the GIAU is the increase in lawsuits and judgments against governmental entities and the increases in insurance rates caused by this increase. We are unable to assess the extent to which the increase in lawsuits is due to claims arising out of activities previously not considered a governmental function, as opposed to claims for which immunity has been waived. Irrespective of the prevalence of claims previously not covered in that increase, the elimination of such claims will remedy the identified concern to some measure by reducing the total number of lawsuits against governmental entities. By extension, that reduction is likely to lower the amount of paid claims and lead to more favorable insurance rates. See id. ¶ 16 (noting that insurance rates “are undoubtedly subject to some measure of fluctuation based on paid claims”). Thus, the means selected is reasonably related to the identified evil.

¶100 As with the other social and economic evils identified, however, the elimination of all claims that would have been previously allowed as a proprietary function is not narrowly tailored. See id. ¶ 17. The amount of paid claims could also be reduced by more narrow measures, including damage caps or the elimination of some, but not all, categories of damages. See id.; see also Craftsman Builder’s Supply, Inc. v. Butler Mfg. Co., 1999 UT 18, ¶ 21, 974 P.2d 1194.32

32 Further, in Laney, the court considered the power company’s ability to increase rates to purchase reasonable amounts of insurance in connection with its open courts clause analysis. See Laney v. Fairview City, 2002 UT 79, ¶ 67, 57 P.3d 1007 (plurality opinion). The District is a “quasi‐municipal” corporation, see Patterick v. Carbon Water Conservancy Dist., 106 Utah 55, 145 P.2d 503, 511 (1944), overruled on other grounds by Timpanogos Planning & Water Mgmt. Agency v. Central Utah Water Conservancy Dist., 690 P.2d 562, 572 (Utah 1984), and while it does not operate at a profit, it receives a substantial portion of its annual budget from the collection of fees.

20100400‐CA 54 ¶101 In sum, the means adopted by the legislature to address the identified social and economic evils are rationally related to those concerns but are not narrowly tailored. Instead of paring back the remedies available to the Jenkinses, the legislature chose to eliminate them entirely. Prior to the adoption of an all‐inclusive definition of governmental function, the District concedes that the Jenkinses could have sued the District for negligence and, if they prevailed, recover the damages caused by the Breaches. The statutory definition of “governmental function” eliminates completely any right of the Jenkinses to recover. Where the remedy selected is not narrowly tailored and no alternative is provided, our supreme court has held that the legislation in question violates the open courts clause, article I, section 11, of the Utah Constitution. See Judd, 2004 UT 91, ¶ 17; Laney v. Fairview City, 2002 UT 79, ¶¶ 69‐71, 57 P.3d 1007 (plurality opinion). Therefore, the GIAU is unconstitutional as applied to the Jenkinses’ claims, and the District cannot take advantage of the immunity that statute purportedly provides.

¶102 In reaching this conclusion, we have attempted to afford great deference to the Utah Legislature’s judgments, while also following the controlling precedent of the Utah Supreme Court concerning our role in interpreting and applying the open courts clause of the Utah Constitution. Meeting both of those obligations is not without its difficulties. Indeed, this analysis highlights the tension between the open courts clause and the separation of powers clause of the Utah Constitution that has caused disagreement among the members of our supreme court. See Utah Const., art. I, § 11; id. art. V, § 1 (“The powers of the government of the State of Utah shall be divided into three distinct departments, the Legislative, the Executive, and the Judicial; and no person charged with the exercise of powers properly belonging to one of these departments, shall exercise any functions appertaining to either of the others, except in the cases herein expressly directed or permitted.”); see also Judd, 2004 UT 91, ¶ 64 (Durham, C.J., and Nehring, J., dissenting) (stating that the damages cap was a “legislative attempt to mandate legal conclusions and clearly violates the separation of powers provision” (internal quotation marks omitted)); Laney, 2002 UT 79, ¶ 85 (Wilkins, J., and Durrant, J., dissenting) (noting that the court’s interpretation of the open courts clause “creates separation of powers problems”). Indeed, one former member of our supreme court reevaluated his position on the substantive nature of the open courts clause due to concern about intrusion on the legislative function. Compare Condemarin v. University Hosp., 775 P.2d 348, 366‐69 (Utah 1989) (Zimmerman, J., concurring in part) (endorsing the Berry test), with Craftsman, 1999 UT 18, ¶¶ 108‐55 (Zimmerman, J., concurring in the result) (calling for the reversal of Berry).

20100400‐CA 55 CONCLUSION

¶103 We remand to the trial court to resolve the factual issue of whether, and the extent to which, the Jenkinses had knowledge of their lost wages and emotional distress damages when they filed their notice of claim. Only those damages that were known but not included in that notice of claim are barred by the GIAU.

¶104 We also hold that the Jenkinses have a special relationship to the District due to their proximity to the Water Line Section that was identified for replacement by the District’s Engineers three years before the first of the Breaches. Thus, the District owed a duty of care to the Jenkinses in determining how long to delay replacing the Water Line Section, and the trial court erred in granting summary judgment in favor of the District under the public duty doctrine.

¶105 The issue of whether the District acted negligently in waiting three years after the Engineers’ recommendation to replace the Water Line Section is within the knowledge and analytical ability of the average layperson. Therefore, the Jenkinses were not required to designate an expert on the standard of care or whether that standard was breached to prove their negligence claim against the District.

¶106 The District’s decisions concerning the order in which Identified Pipe would be replaced, the amount of money allocated to capital improvements, and the coordination of replacements with road projects of the relevant municipalities involved the expertise and policy decisions of the District. Thus, the decision of when to replace the Water Line Section was a discretionary function entitled to immunity under the GIAU.

¶107 The District concedes that the all‐inclusive definition of governmental function contained in the 1987 amendments to the UGIA, as well as the 2004 enactment of a substantially similar definition of governmental function, resulted in the abrogation of the Jenkinses’ preexisting remedy. We conclude that the legislature did not provide the Jenkinses any reasonable alternative remedy. While the adoption of the “governmental function” definition is logically designed to address, to some measure, the social and economic evils identified by the legislature, it eliminates all claims for damages caused by the District’s negligence in the provision of water, irrespective of the amount or character of those damages. As a result, the means employed by the legislature to eliminate the social and economic evils it has identified are not narrowly tailored. Therefore, the definition is unconstitutional as applied here because it violates the

20100400‐CA 56 Jenkinses’ rights under the open courts clause of article I, section 11 of the Utah Constitution. The District may not rely on the GIAU as a basis for immunity from liability to the Jenkinses for its negligence, if any, in deferring the replacement of the Water Line Section.

¶108 We reverse the summary judgment in favor of the District and remand for trial on the Jenkinses’ negligence claims in accordance with this decision.

____________________________________ Carolyn B. McHugh, Presiding Judge

¶109 WE CONCUR:

____________________________________ J. Frederic Voros Jr., Associate Presiding Judge

____________________________________ Gregory K. Orme, Judge

20100400‐CA 57