TEXAS DEPARTMENT OF INSURANCE, Honorable Mike Geeslin, Commissioner of Insurance and Honorable Danny Saenz, Senior Associate Commissioner, Petitioners, v. AMERICAN NATIONAL INSURANCE COMPANY and American National Life Insurance Company Of Texas, Respondents, 410 S.W.3d 843

Justice MEDINA delivered the Opinion of the Court. The issue in this appeal is whether stop-loss insurance sold to self-funded employee health-benefit plans is “direct health insurance” or “reinsurance.” The distinction is significant because direct insurance is subject to state insurance regulation, while reinsurance is not. Reinsurance is not regulated because it typically involves the reallocation of risk between two insurance companies rather than a consumer-insurance transaction.

The parties to this appeal disagree about whether an employer who self funds a health-benefit plan for its employees is an “insurer” under the Texas Insurance Code, and therefore should be treated as a reinsurer when purchasing stop-loss insurance. The court of appeals concluded that an employer’s self-funded plan was clearly an insurer under the Texas Insurance Code and that a plan’s purchase of stop-loss insurance was also clearly reinsurance beyond the regulatory scope of the Texas Department of Insurance. 364 S.W.3d 299, 301 (Tex.App.-Austin 2010) (mem.op.).

The court accordingly reversed the trial court’s judgment, which had sustained the agency’s regulation of the stop-loss policies at issue as direct insurance. Because the regulatory agency did not clearly err in its regulation of these stop-loss policies, however, we reverse the court of appeals’ judgment and render judgment for the agency. I American National Insurance Company and American National Life Insurance Company of Texas (collectively American) are licensed to sell insurance in Texas. American sells stop-loss insurance to self-funded employee health-benefit plans, among other types of policies.

Under a self-funded benefit plan, an employer assumes the risk of providing health insurance to its employees, instead of ceding the risk to a third-party insurance company. The employer then either sets aside funds for its employees’ covered medical expenses or pays for such expenses out of its general accounts. Self-funded plans typically hire third parties to administer the plan and often purchase stop-loss insurance to limit financial exposure to catastrophic losses.

During a routine audit, the Texas Department of Insurance discovered that American had sold stop-loss policies between January 1998 and December 2002 without paying taxes or complying with other regulatory requirements applicable to insurers. The Department later formally found that American had violated

article 3.10(

a) of the Insurance Code by “improperly recording the direct stop-loss policy premiums obtained from the self-insured employers as ‘assumed reinsurance,’ ” rather than as “direct written premium.” The Department reasoned that, because self-funded employers to which American sold its stop-loss policies were not themselves “insurers authorized to do the business of insurance,” stop-loss coverage was not “assumed reinsurance.” The Department further found that American had failed to pay assessments due the Texas Health Insurance Risk Pool on these stop-loss policies in violation of

article 3.77. Finally, the Department found that American had failed to submit these policy forms to the Department for approval or to request an exemption as required by the Administrative and Insurance Codes. See 28 Tex. Admin. Code §§ 3.4002, 3.4004(e)(2)(J), and Tex. Ins.Code art. 3.42 (repealed). After exhausting its administrative remedies, American sued the Department, seeking declaratory and injunctive relief. American contended that its stop-loss policies were reinsurance over which the Department lacked regulatory authority.

It asked the trial court to declare the Department’s actions invalid and to enjoin the Department from enforcing its findings. The Department, on the other hand, argued that American’s stop-loss policies were direct insurance subject to the Texas Insurance Code and its regulatory authority. Both American and the Department filed motions for

summary judgment. The trial court granted the Department’s motion and denied American’s, causing American to appeal. In a memorandum opinion, the court of appeals concluded that self-funded employee health-benefit plans were insurers under Texas law and that the stop-loss policies sold to the plans by American were therefore reinsurance rather than direct insurance. 364 S.W.3d at 301. Moreover, the court concluded that the Department’s contrary view was entitled to no deference because such view was plainly inconsistent with the Insurance Code. Id. at 307.

The court of appeals accordingly reversed the trial court’s judgment, holding the Department’s findings of Insurance Code violations to be invalid because American’s stop-loss policies, as reinsurance, were not subject to the Department’s regulation. Id. at 308. II American contends that an employer who self funds a health-benefit plan for its employees is an “insurer” in the “business of insurance” under the Insurance Code and therefore a reinsurer when purchasing stop-loss insurance.

According to American, the plan’s purchase of stop-loss insurance is a redistribution of the risk assumed by the plan in the same sense as a reinsurance contract is a redistribution of risk from one insurance company to another. Because reinsurance contracts are not subject to regulation under the Insurance Code, American concludes that the Department erred in categorizing its stop-loss policies as direct health insurance and requiring it to comply with those provisions applicable to the sale of that kind of insurance.

The Department responds that reinsurance is the redistribution of risk between sophisticated insurers in the business of insurance and that an employee health-benefit plan is neither as a matter of law. Although an employee health-benefit plan may in some ways act like an insurer with respect to the plan’s participants, the Insurance Code does not regulate it as one. Insurance purchased by the plan is therefore not reinsurance, according to the Department.

It is instead direct insurance in the nature of health insurance because the stop-loss policies are purchased by the plans to cover ultimate claims associated with their health-care expenses. As the court of appeals acknowledged, the term “reinsurance” is not defined in the Insurance Code. Moreover, the term’s common meaning has become confused over time by indiscriminate use. As one authority has commented: The term “reinsurance” has been used by courts, attorneys, and writers with so little discrimination that much confusion has arisen as to what that term actually connotes.

Thus, it has so often been used in connection with transferred risks, assumed risks, consolidations and mergers, excess insurance, and in so many other connections that it now lacks a clean-cut field of operation. 1 ERic Mills Holmes & Mark S. Rhodes, Holmes’ Appleman on Insuranoe § 2.15, at 317 (2d ed.1996). The Insurance Code also does not define stop-loss insurance. Stop-loss insurance, however, has similarities with both excess insurance and reinsurance.

See Kenneth Thompson, Reinsur-anoe 8 (4th ed. 1966) (“The self-insurer for economic stability and soundness will obtain coverage which some may call ‘reinsurance’ but which is really of a type of ‘specific’ excess insurance.”). Under a stop-loss policy, the insurer agrees to reimburse a self-funded plan for healthcare costs that exceed a contractually predetermined amount. The obligation generally takes one of two forms: specific or aggregate.

Specific stop-loss policies cover claims over a certain dollar amount per employee, while aggregate stop-loss policies provide a cap for an employer’s overall liability for all covered persons. Similarly, excess insurance is an agreement to indemnify against any loss that exceeds the amount of primary or other coverage. Nat’l Union Fire Ins. Co. of Pittsburgh, Pa. v. Ins. Co. of N. Am., 955 S.W.2d 120, 138 (Tex.App.-Houston [14th Dist.] 1997), aff'd sub nom. Keck, Mahin & Cate v. Nat’l Union Fire Ins. Co. of Pittsburgh, Pa., 20 S.W.3d 692 (Tex.2000).

Reinsurance, on the other hand, has been described as the transfer of all or part of one insurer’s risk to another insurer, which accepts the risk in exchange for a percentage of the original premium. Stark v. Geeslin, 213 S.W.3d 406, 410 n. 2 (Tex.App.-Austin 2006, no pet.) (citing Black’s Law Dictionary 1290 (7th ed.1999)). “The true reinsurer is merely an insurance company or underwriter which deals only with other insurance companies as its policyholders.” Great Atl. Life Ins. Co. v.

Harris, 723 S.W.2d 329, 330 (Tex.App.-Austin 1987, writ dism’d) (quoting Appleman, Insurance Law and Practice § 7681, at 480 (1976)). Both direct insurance and reinsurance reallocate risk with the principal distinction being the nature of the purchaser. Insurance consumers reallocate their risk by purchasing direct insurance, both primary and excess, while insurance companies reallocate the risks they assume by purchasing reinsurance. Employers who self fund their employee health-benefit plans are clearly not insur-anee companies, but they perform a similar service.

This was enough for the court of appeals to conclude that the self-insured employer is an insurer. The court observed that self-funded plans are insurers under Texas law because they meet the broad definition of “insurer” found in the Insurance Code

chapter prohibiting the unauthorized business of insurance, and because much of what the plans do comports with that same chapter’s definition of conduct constituting the “business of insurance.” See 364 S.W.3d at 304 (quoting Tex. Ins.Code §§ 101.002(1)(A), 101.051). The

definitions used by the court of appeals are from

Chapter 101 of the Insurance Code. The chapter’s stated purpose is to “subject certain insurers and persons to the jurisdiction of the [insurance] commissioner and proceedings before the commissioner” and “the courts of this state in suits by or on behalf of the state or an insured or beneficiary under an insurance contract.” Tex. Ins.Code § 101.001(c)(1), (2). The legislative concern was that insurers were operating in this state without authorization and, therefore, were evading the Code’s regulatory framework. Id. § 101.001(a), (b)(4).

The Legislature therefore broadly defined the terms “insurer” and “business of insurance” to capture all unauthorized activity. The court of appeals use of the chapter’s

definitions here is somewhat ironic because it removes American’s stop-loss insurance from the state’s regulatory authority. Without question, self-funded employee health-benefit plans operate much like insurers. Their activities not surprisingly then fit the

definitions of “insurer” and “business of insurance” found in the

chapter designed to prohibit the unauthorized business of insurance. But that chapter’s purpose is to extend the state’s regulatory authority to those conducting the business of insurance in the state without authorization. That purpose does not include self-funded employee health-benefit plans because they are not regulated like insurance companies. Most private self-funded plans qualify as “employee welfare benefit plans” under the federal Employee Retirement Income Security Act (ERISA). 29 U.S.C. § 1002(1).

ERISA prohibits states from deeming these self-funded plans “insurance companies] or other insurer[s]” or “to be engaged in the business of insurance” for purposes of state insurance regulation. Id. § 1144(b)(2)(B). Simply put, states cannot regulate private self-funded insurance plans. Id. While

Chapter 101 defines insurer and the business of insurance, it does so only for purposes of that particular chapter. Id. § 101.002. These

definitions are not comprehensive or for all purposes as the Insurance Code provides variations in different contexts. For example, the

article that governs insurance carrier licensing defines “insurer” as “the issuer of an insurance policy that is issued to another in consideration of a premium and that insures against a loss that may be insured against under the law.” Tex. Ins.Code § 801.001(2) (formerly Tex. Ins.Code art. 1.14, § 2). This definition excludes self-funded employers, who do not issue policies for a premium.

Although the Department contends that the Legislature has left it to define the difference between stop-loss insurance and reinsurance, it submits that this would have been a more sensible definition for the court of appeals to have used in its analysis. As we have previously observed, the “Insurance Code is somewhat different from Texas’s other statutory codifications in that it is not a formal, unified Code containing uniform

definitions.” Dallas Fire Ins. Co. v. Tex. Contractors Sur. & Cas. Agency, 159 S.W.3d 895, 896 (Tex.2004) (per curiam) (citing Great Am. Ins. Co. v. N. Austin Mun. Util. Dist. No. 1, 908 S.W.2d 415, 424 (Tex.1995)). In Dallas Fire, we considered a similar question regarding the contextual application of Insurance Code

definitions. There, the issue was whether former

article 1.14-1’s definition of “business of insurance” applied equally to

article 21.21, which provides a private cause of action for unfair or deceptive acts in the “business of insurance.” Dallas Fire Ins. Co., 159 S.W.3d at 895. A surety company argued that it was not in the business of insurance as used in

article 21.21, despite the fact that former

article 1.14-1 defined the business of insurance to include the business of suretyship. Id. at 896. Former

article 1.14-1 was the predecessor to the unauthorized insurance provisions now found in

Chapter 101 and contained essentially the same “business of insurance” definition. Indeed, the language of the specified conduct constituting the business of insurance at issue in Dallas Fire Insurance has not changed. Compare Tex. Ins.Code art. 1.14-1, § 2(a) (repealed), with Tex. Ins.Code § 101.051. Former

article 1.14-1, like

Chapter 101, provided that, among other conduct, “[t]he making of or proposing to make, as guarantor or surety, any contract of guaranty or suretyship as a vocation and not merely incidental to any other legitimate business or activity of the guarantor or surety” constituted the insurance business. Tex. Ins.Code art. 1.14-1, § 2(a)(2). In concluding that the business of insurance for purposes of

article 21.21 did not include suretyship, we relied on Great American Insurance Company v. North Austin Utility District No. 1, where we said: Nowhere in the “purpose” clause of

article 1.14-1 did the Legislature indicate that the list of acts contained therein which constitute “doing an insurance business” was to apply throughout the Code. Rather, the purpose clause of

article 1.14-1 points out that in defining “what constitutes doing an insurance business,” the Legislature was exercising its power to address its explicitly listed concerns. The expressed concerns do not evidence an intention to promulgate a uniform definition of the acts which constitute doing an insurance business; rather, they indicate concern that particular parties may escape the jurisdiction of the State Board of Insurance and evade suit by contractual beneficiaries. Great Am., 908 S.W.2d at 423 (emphasis added).

Applicable here is our observation that “ ‘the business of insurance’ has meant different things in different sections of the Code.” Dallas Fire Ins. Co., 159 S.W.3d at 896. Thus, unlike the court of appeals, we do not find

Chapter 101’s

definitions to be determinative in this case, and we must look elsewhere for guidance. The Department submits that we should look to the statutes at issue in this case for the answer. Ill For more than a decade, the Department has categorized stop-loss coverage as direct insurance (not reinsurance) subject to assessment by the Texas Health Insurance Risk Pool under

article 3.77 and subject to the reporting requirements of

article 3.42. From 1998 to 2002, former

article 3.77 authorized the Pool to “assess insurers ... for the [Pool’s] organizational and interim operating expenses.” Tex. Ins. Code art. 3.77, § 13(a). For purposes of this assessment, the Code defines “insurer” as: [A]ny entity that provides health insurance in this state, including stop-loss or excess loss insurance. For the purposes of this article, “insurer” includes but is not limited to an insurance company; ... an insurer providing stop-loss or excess loss insurance to ... any benefit arrangements to the extent permitted by [federal law].... Id. art. 3.77, § 2(11) (emphasis added).

The Department argues that by including stop-loss insurance in the above definition the Legislature intended to include it as a type of health insurance subject to assessment by the Pool. The Department further submits that

article 3.77’s definition of “health insurance” is to the same effect. Health insurance is defined to include “any hospital and medical expense incurred policy ... or any other health care plan or arrangement that pays for or furnishes medical or health care services whether by insurance or otherwise.” Id. art. 3.77, § 2(7). Because stop-loss policies are designed to cover claims for hospital or medical expenses that exceed a predetermined attachment point, either individually or in the aggregate, the Department maintains that the policies qualify as a “hospital and medical expense incurred policy” subject to assessment. Soon after the passage of

article 3.77, the Department proposed a regulation instructing that stop-loss and excess policies like American’s were subject to assessment. See 28 Tex. Admin. Code § 3.13001 (1999) (Tex. Dep’t of Ins.). The rule was open for public comment for 30 days before adoption but no one questioned the rule’s explicit inclusion of stop-loss and excess-loss policies. See 23 Tex. Reg. 1309, 1311 (1998). Because this regulation is reasonable and consistent with

article 3.77, the Department argues that it was entitled to deference. First Am. Title Ins. Co. v. Combs, 258 S.W.3d 627, 632 (Tex.2008). American argues, however, that the Fifth Circuit has already determined under Texas law that stop-loss insurance sold to self-insured employee health-benefit plans is not direct health insurance. See Brown v. Granatelli, 897 F.2d 1351, 1354 (5th Cir.1990) (concluding “that under Texas law stop-loss insurance is not accident and sickness insurance”). Brown involved former

article 3.70-2(E), which required that accident-and-siekness insurance cover newborns from birth. TEX. INS. CODE art. 3.70-2(E) (repealed). The underlying plan was an employee welfare-benefit plan within the meaning of ERISA. Brown, 897 F.2d at 1352. The plan excluded coverage for all newborns during the child’s first 31 days and thereafter for any child who was disabled, hospitalized, or sick on the 31st day. Id. at 1353. The employer, however, also purchased stop-loss insurance to reimburse the plan for claims paid that exceeded $30,000 for anyone covered during the policy year. Id.

The plan refused to pay for medical costs associated with two premature births, and the parents sued the employer and the plan, who joined the stop-loss insurer as a third-party defendant. The parents admitted that ERISA prohibited the direct application of

article 3.70-2(

E) to the plan, but argued that under Metropolitan Life Insurance Co. v. Massachusetts, 471 U.S. 724, 105 S.Ct. 2380, 85 L.Ed.2d 728 (1985), the provision could be applied to the plan indirectly through the stop-loss policy. The stop-loss insurer argued that an insurance policy purchased by an employee-benefit plan to protect the plan from catastrophic loss was not accident-and-sickness insurance even though it indirectly covered accident and sickness losses. Neither argument attracted a majority of the court, although a majority did agree that the stop-loss insurer was not liable.

The court’s decision produced three separate opinions. The lead opinion reasoned that the stop-loss policy was primary coverage for the plan’s catastrophic losses rather than the individual losses of the group’s participants and therefore not accident-and-sickness insurance to which the statute applied. Brown, 897 F.2d at 1354 (Higginbotham, J.).

The lead opinion cautioned in dicta, however, that a stop-loss policy might be an accident-and-sickness policy if coverage were to trigger at an unreasonably low amount: “If, for example, a plan paid only the first $500 of a beneficiaries’ health claim, leaving all else to the insurer, labeling its coverage stop-loss or catastrophic coverage would not mask the reality that it is close to a simple purchase of group accident and sickness coverage.” Id. at 1355.

The dissenting justice complained that this distinction was “contrary to the substance of the Texas Insurance Code, and unworkable as a standard for future cases.” Id. at 1356 (Brown, J., dissenting). The dissent agreed that ERISA preempted state law from regulating the group plan’s content but did not agree that it also preempted regulation of the insurance company that sold the stop-loss insurance to the plan. Id. at 1357.

The dissent concluded that ERISA’s “insurance savings clause” left the stop-loss insurer subject to state insurance law, despite the preemption of that law as to the plan, under the Supreme Court’s reasoning in Metropolitan. Id. (citing Metropolitan, 471 U.S. at 740-41, 105 S.Ct. 2380). The third and final justice in Brown agreed with both of his colleagues that ERISA preempted state law from requiring the benefit plan to cover newborns, and therefore the plan could not incur losses for such claims. Id. at 1355 (Reavley, J., concurring).

Because the stop-loss insurer had agreed to reimburse the plan only for losses exceeding $30,000 and that trigger had apparently not been reached, the third justice concurred in the judgment affirming the

summary judgment for the plan and its stop-loss insurer. ERISA’s broad preemption provision provides that: Except as provided in subsection (

b) of this section, the provisions of this sub-chapter and subchapter III of this

chapter shall supersede any and all State laws insofar as they may now or hereafter relate to any employee benefit plan described in

section 1003(

a) of this title and not exempt under

section 1003(

b) of this title. 29 U.S.C. § 1144(a), ERISA § 514(a). This preemption provision, however, is modified by

section 514(b), the “insurance savings clause,” which provides in pertinent part: Except as provided in subparagraph (B), nothing in this subchapter shall be construed to exempt or relieve any person from any law of any State which regulates insurance, banking, or securities. 29 U.S.C. § 1144(b)(2)(A). Subparagraph (

B) is the “deemer clause” that exempts plans from the operation of state laws regulating insurance. In Metropolitan, the Supreme Court concluded ERISA preempted the direct application of mandated-benefit laws to employee-benefit plans, but it did not necessarily preempt their indirect application to the insurance policies purchased by such plans. 471 U.S. at 736 & n. 14, 747 & n. 25, 105 S.Ct. 2380. In the context of that case, the Court observed that ERISA would cause insured and uninsured plans to receive different treatment. Id. at 747, 105 S.Ct. 2380.

This distinction subsequently led courts to struggle with the question of whether an employee health-benefit plan that self insures its members but also purchases stop-loss insurance is functionally insured or self-insured. Such plans would appear to be a bit of both. Some courts, like the lead opinion in Brown, have suggested that the answer lies in whether the plan is predominantly insured or self-insured, but that approach has been criticized as fundamentally misguided. See Russell Korobkin, The Battle Over Self-Insured Health Plans, or “One Good Loophole Deserves Another,” 5 Yale J.

Health Pol’y L. & Ethics 89, 115 (2005). Unlike Brown, the question in this case does not involve the extent of coverage under the group health-benefit plan, either directly or indirectly, or the contractual relationship between a plan and its stop-loss insurer. Instead, the questions are (1) whether the state can regulate stop-loss insurers who contract with such plans, as it does other direct health-care insurers by requiring them to contribute to the Pool and to submit their policies for approval, and (2) whether it has chosen to do so.

The answer to the first question is clearly yes under ERISA’s “insurance savings clause” and the Supreme Court’s decision in Metropolitan. The answer to the second question is less clear, but the Department’s longstanding

interpretation of the statute is entitled to serious consideration. IV Our primary objective when construing a statute is to determine and give effect to the Legislature’s intent. McIntyre v. Ramirez, 109 S.W.3d 741, 745 (Tex.2003). We begin with the words the Legislature used because the surest guide to what lawmakers intended is what they enacted. First Am. Title Ins. Co., 258 S.W.3d at 631. “ ‘If the statute is clear and unambiguous, we must apply its words according to their common meaning’ in a way that gives effect to every word, clause, and sentence.” Id. (quoting State v.

Shumake, 199 S.W.3d 279, 284 (Tex.2006)). But when a statute’s meaning is ambiguous, we frequently defer to administrative agencies’ statutory

interpretations. “Construction of a statute by the administrative agency charged with its enforcement is entitled to serious consideration, so long as the construction is reasonable and does not contradict the plain language of the statute.” Tarrant Appraisal Dist. v. Moore, 845 S.W.2d 820, 823 (Tex.1993). This is particularly true where the agency’s

interpretation has been sanctioned by long acquiescence. Stanford v. Butler, 142 Tex. 692, 181 S.W.2d 269, 273 (1944). If a statute is vague or ambiguous, we defer to the agency’s

interpretation unless it is plainly erroneous or inconsistent with the language of the statute. TGS-NOPEC Geophysical Co. v. Combs, 340 S.W.3d 432, 438 (Tex.2011). Our deference, however, is “tempered by several considerations.” R.R. Comm’n of Tex. v. Tex. Citizens for a Safe Future & Clean Water, 336 S.W.3d 619, 625 (Tex.2011). An administrative agency’s construction of a statute it implements ordinarily warrants deference when: (1) the agency’s

interpretation has been formally adopted; (2) the statutory language at issue is ambiguous; and (3) the agency’s construction is reasonable. Id. (quoting Fiess v. State Farm Lloyds, 202 S.W.3d 744, 747-48 (Tex.2006)). The Legislature has chosen not to define the terms “stop-loss insurance” and “reinsurance” in the Insurance Code. American, however, identifies provisions in the Code where the term “reinsurance” is used in connection with self-funded plans in support of its argument that stop-loss insurance is reinsurance. See, e.g., Tex.

Gov’t Code § 845.406(f) (to protect against adverse claim experience, a self-funded state employee benefit arrangement may secure reinsurance); Tex. Ins.Code § 1551.208 (under Texas Employees Group Benefits Act, the board of trustees may reinsure any coverage the board decides to self-insure out of its life, accident, and health insurance fund). American contends these statutes express the Legislature’s view of stop-loss insurance as reinsurance, which is beyond the Department’s regulatory authority. See id. § 101.053(b)(2).

The Department responds that the Legislature has used the terms reinsurance and stop-loss insurance generally as shorthand for the redistribution of risk without defining them in any specific or technical sense. See, e.g., Tex.

Ins.Code § 846.053(h) (requiring multiple employer welfare arrangements to purchase specific and aggregate stop-loss insurance); id. § 1506.107 (authorizing pool to “contract for stop-loss insurance for risks incurred by the pool”); id. § 1508.261 (providing the fund “may purchase stop-loss insurance or reinsurance from an insurance company licensed to write that coverage in this state”); id. § 1506.002(a)(3) (defining a health care plan or arrangement as “including stop-loss insurance or excess loss insurance or reinsurance for individual or group health insurance or for any other health care plan or arrangement”).

Because the Insurance Code does not define these terms or use them consistently, the parties are left to emphasize the provisions most favorable to their respective

interpretations. Those provisions yield competing plausible

interpretations but no definitive answer under the Code. We conclude then that the Insurance Code is ambiguous on how stop-loss insurance should be treated. To fill the void, the Department submits that it has promulgated a rule instructing that stop-loss and excess loss policies like American’s are in the nature of direct health insurance, not reinsurance, and subject to assessment under former

article 3.77. The Department reasons that reinsurance is the redistribution of risk between sophisticated insurers in the business of insurance and that an employee health-benefit plan is neither as a matter of law. Although an employee health-benefit plan may in some respects act like an insurer with respect to the plan’s participants, the Insurance Code does not regulate it as one, and ERISA generally precludes the Code from deeming these plans to be insurers or in the business of insurance. 29 U.S.C. § 1144(a); FMC Corp. v. Holliday, 498 U.S. 52, 61, 111 S.Ct. 403, 112 L.Ed.2d 356 (1990).

The Department has therefore concluded that stop-loss insurance purchased by a plan does not involve two insurers and is therefore not reinsurance. It is instead direct insurance in the nature of health insurance because the stop-loss policies are purchased by the plans ultimately to cover claims associated with their health-care expenses. The Department’s construction is reasonable, was formally promulgated, and is not expressly contradicted by the Insurance Code.

We accordingly agree with the Department’s construction and hold that stop-loss insurance sold to a self-funded employee health-benefit plan is not reinsurance, but rather direct insurance subject to regulation under the Insurance Code. The court of appeals’ judgment is reversed, and judgment is rendered for the Department. Justice HECHT did not participate in the decision. . See Act of June 14, 1995, 74th Leg., R.S., ch. 614, § 2, 1995 Tex. Gen. Laws 3468, 3468-69, repealed by Act of June 16, 2005, 79th Leg., R.S., ch. 727, § 18(a)(3), 2005 Tex. Gen. Laws 1752, 2186-87. .

See Act of June 16, 1989, 71st Leg., R.S., ch. 1094, § 2, 1989 Tex. Gen. Laws 4484, 4484-91, repealed by Act of June 21, 2003, 78th Leg., R.S., ch. 1274, § 26(a)(1), 2003 Tex. Gen. Laws 3611, 4138. .See Act of May 23, 1995, 74th Leg., R.S., ch. 176, § 1, 1995 Tex. Gen. Laws 1889, 1889-92, repealed by Act of June 21, 2003, 78th Leg., R.S., ch. 1274, § 26(a)(1), 2003 Tex. Gen. Laws 3611, 4138. . During part of the relevant period these

definitions were located in

article 1.14-1 of the Code. See Tex. Ins.Code art. 1.14-1, § 2(a) (repealed). .ERISA does not apply to self-funded plans of governmental entities and churches, but state law exempts those entities from regulation as well. See 29 U.S.C. § 1003(b)(1), (2); Tex. Bus Orgs.Code § 22.409 (exempting church plans); Tex. Gov't Code § 2259.037 (exempting governmental plans). . See Act of June 7, 1951, 52nd Leg., R.S., ch. 491, § 1, art. 1.14, 1951 Tex. Gen. Laws 868, 872, repealed by Act of May 22, 2001, 77th Leg., ch. 1419, § 31(a), 2001 Tex. Gen. Laws 3658, 3662. .

Act of May 28, 1987, 70th Leg., R.S., ch. 254, § 1, 1987 Tex. Gen. Laws 1573, 1573, repealed by Act of May 17, 1999, 76th Leg., R.S., ch. 101, § 1, 1999 Tex. Gen. Laws 486, 525-26. . The "Risk Pool is a quasi-governmental entity [created by the Legislature] to provide affordable insurance to Texans who have preexisting conditions or other high-risk conditions that might prevent them from obtaining insurance otherwise.” Tex. Health Ins. Risk Pool v. Sigmundik, 315 S.W.3d 12, 13 n. 1 (Tex.2010). .

The Department similarly argues that stop-loss coverage as a form of health insurance was also subject to the reporting requirements of Texas Insurance Code

article 3.42. . The Department further submits that its

interpretation has been subsequently confirmed by the Legislature's 2003 clarification that the term health insurance includes "stop-loss insurance or excess loss insurance or reinsurance.” See Act of May 29, 2003, 78th Leg. R.S., ch. 840, § 1, 2003 Tex. Gen. Laws 2627, 2627. . "[W]hether an EHBP maintains the actual insurance risk associated with employee illness bears no direct relevance to the question of whether the deemer clause, according to its text, prohibits state regulation of its members' health insurance contracts.

Courts need only ask which entity promises to pay the health care costs incurred by plan members. If the EHBP must pay these costs, and thus acts as an insurer of its employee’s health care, the state may not regulate the provisions of the employee-EHBP contract, and the plan is therefore 'self-insured' according to the Metropolitan Life dichotomy. If a third-party insurance company bears the insurance risk of the employee’s health care, the state may regulate the insurance contract, and the plan is therefore 'insured' under Metropolitan Life.

Whether a self-insured plan does or does not purchase stop-loss insurance, or whether that stop-loss insurance has a low or high attachment point, is simply irrelevant, at least under a close reading of ERISA’s text.” . The court of appeals here remarked that a court would be bound by an agency’s reading of a statute, if reasonable and in harmony with the rest of the statute. 364 S.W.3d at 303 quoting City of Plano v. Pub. Util. Comm'n, 953 S.W.2d 416, 421 (Tex.App.-Austin 1997, no writ).

Although we have never referred to an agency's reasonable and harmonious reading of a statute as binding on the Court, we have acknowledged the agency’s

interpretation to be worthy of "serious consideration” and "great weight.” See, e.g., Mid-Century Ins. Co. of Tex. v. Ademaj, 243 S.W.3d 618, 623 (Tex.2007); Osterberg v. Peca, 12 S.W.3d 31, 51 (Tex.2000).