dissenting:
Because the majority opinion (1) disregards our holding in Robertson v. Alexander Grant & Co., 798 F.2d 868 (5th Cir.1986) to find that the insurance policy is governed by ERISA and (2) ignores the provisions of Louisiana Revised Statute section 22:230 and Louisiana jurisprudence on total disability policy definitions to conclude that Walter House does not qualify as totally disabled under the policy language, I respectfully dissent.
I. ERISA Coverage
As an initial matter, the majority misstates the standard of review for the question of whether House’s insurance policy constitutes an ERISA plan. In McNeil v. Time Insurance Co., 205 F.3d 179, 181 (5th Cir.2000), we reviewed an identical question in a similar posture. There, a district court had dismissed state law claims on summary judgment on the grounds that a single insurance policy covering both a partner and an employee as beneficiaries constituted an ERISA plan, preempting the state law claims. Id. at 189. We applied two different standards: the question of whether the “insurance policy constituted an ERISA plan” was a fact issue reviewed for clear error, while the question of whether ERISA preempted state law claims was a legal one reviewed de novo. Id.; see also Provident Life and Acc. Ins. Co. v. Sharpless, 364 F.3d 634, 638 (5th Cir.2004). Here, the majority recasts the same question of whether the insurance policy constitutes an ERISA plan as a mixed question of fact and law that is subject to de novo review. Applying McNeil, we should review the district court’s decision that the partner House’s policy was separate from the employees’ policy for clear error, and I do not bélieve that any error here is significant enough to overcome that hurdle.
Under either standard, however, I would affirm the district court’s determination that House’s policy was not part of an ERISA plan. In Robertson, we held that plans benefiting only partners were not covered by ERISA because of the unique negotiating posture a partner assumes visa-vis his firm. Robertson noted that
[e]mployees in the traditional employer-employee relationship are more vulnerable than partners in a partnership are to abuses because workers typically lack control over pension plan management and input into the decision whether to extend pension benefits to certain employees. On the other hand, a partner has more control and input than does an employee since a partner has a vote in partnership affairs. Furthermore, a partnership contains a self-policing feature largely absent in the typical employer-employee relationship. In the *457partnership situation the partners have an incentive not to agree to provisions that may harm certain members of the partnership because each partner knows that he could end up being the partner who is harmed.
798 F.2d at 870.
This logic is especially forceful here. House was not only a partner of the law firm, but a founder with his name in the most prominent position. House’s superi- or control and input as a partner is evident from the facts found by the district court. It noted that House was able to obtain advantages over the policy given to employees. House’s income covered by the policy included his entire net income, whereas employees were limited to their base salary under a forty hour workweek. House v. Amer. United Life Ins. Co., No. 02-1342, 2004 WL 856671, at *9 (E.D.La. Apr.20, 2004). House also was able to secure a much more favorable definition of total disability as compared to his employees. He could be considered totally disabled if “because of Injury or Sickness the Person cannot perform the material and substantial duties of his regular occupation,” whereas employees could be totally disabled only if they “cannot perform the material and substantial duties of any gainful occupation for which the Person is reasonably fitted by training, education, or experience.” Id. (emphasis added). Under Robertson’s rationale, House was not vulnerable to abuse under his policy because of his status as a partner and ERISA was not intended to provide him with protection.
Robertson does not support the contention that if the policy is “intertwined” with policies of employees, a partner’s separate policy becomes covered by ERISA. To the contrary, Robertson rejected an argument that the plans in that case were virtually identical and thus were really a single plan. The intertwinement argument “ignores the fact that the plans, however similar, are two separate plans. The plan covering the partners does not pay any benefits to principals, and the plan covering principals does not pay any benefits to partners.” Robertson, 798 F.2d at 871. If two virtually identical plans covering different groups of employees are considered separate plans, I see no basis here to hold that there is a single policy where partners paid for their policies individually while the premiums of employees were paid by the firm, partners received different benefits from those negotiated for employees, and employees were required to participate whereas partners were not. That the policies were purchased at the same time and both tasked the firm with minimal administrative duties does not adequately distinguish the case from Robertson in my view. Robertson is vague as to its facts, but appears to involve two parallel and separate Keogh plans, with nearly identical terms, administered by an accounting partnership on behalf of its partners and its employees.
There is certainly nothing about purchasing policies for employees and partners at the same time that eliminates the bargaining advantage a partner has over an employee. To the contrary, it likely enhances it — the partners, able to control which insurance company their employees are forced to select, gain added negotiating leverage for their own policies. Likewise, common administration of separate and distinct policies for employees and employers does not disadvantage the partner in any way — House as a partner was in a prime position to control the administration of his own policy. Robertson reasoned that because partners were situated such that they could exercise control over their own plan, the application of ERISA to partners was unnecessary to protect *458them from their employers — in other words, themselves. Even if the policies here were intertwined, that does not change the fact that House as a partner did not need protection under ERISA, and that, under the plain terms of the statute, protection would not extend to his policy unless an employee was covered by it. And here, as Robertson noted, “the [policies], however similar, are two separate [policies].” Id.
Other Circuits are in accord. The Ninth Circuit has decided this issue and taken the opposite approach from the majority opinion. In In re Watson, 161 F.3d 593, 596 n. 4 (9th Cir.1998), a partner argued that while his plan was separate from that of the employees, it was sufficiently related to the employee’s plan as to warrant ERISA coverage for his own plan as well. The court rejected this argument, holding that “even if the plans were created simultaneously or shared other common characteristics, they are independent plans under ERISA.” Id.; see also LaVenture v. Prudential Ins. Co. of Amer., 237 F.3d 1042, 1046 (9th Cir.2001) (reiterating Watson and adding that “a company may offer more than one benefit plan, one covering only the owner of the business and the other covering the business’s employees, and maintain those two plans as independent plans under ERISA”).
Similarly, the Eleventh Circuit has held that “non-ERISA benefits do not fall within ERISA’s reach merely because they are included in a multibenefit plan along with ERISA benefits.” Kemp v. IBM Corp., 109 F.3d 708, 713 (11th Cir.1997). Kemp noted that including a non-ERISA benefit within an ERISA plan does not convert that benefit into one covered by ERISA. Id. Judge Livavdais of the Eastern District of Louisiana, in applying Robertson to reach a holding similar to that in Kemp, noted the potential consequences of any other interpretation:
If any court endorsed this view, the insured’s choice to participate in a policy other than a group policy would be rendered meaningless; ERISA would become a boundless piece of legislation, sweeping within its scope all claims made under separate policies purchased to provide benefits in addition to those provided by ERISA plans.
St. Martin v. Provident Life & Acc. Ins. Co., Nos. 92-2120, 92-4244, 1993 WL 262708 (E.D.La. July 2, 1993).
Moreover, the cases cited by the majority do not change my view that Robertson should guide the Court’s analysis here. As the majority points out, Yates dealt with situations in which an employee benefit plan covered both a working owner, such as a partner, and at least one employee. See Raymond B. Yates, M.D., P.C. Profit Sharing Plan v. Hendon, 541 U.S. 1, 124 S.Ct. 1330, 158 L.Ed.2d 40 (2004). The Supreme Court specifically noted that “if a benefit plan covers only working owners, it is not covered by Title I [of ERISA].” Id. at 22 n. 6, 124 S.Ct. 1330 (citing appellate cases) (emphasis added); see also id. at 21, 124 S.Ct. 1330 (“Plans that cover only sole owners or partners ... fall outside [ERISA’s] domain.”). Thus, Yates does not support the majority’s intertwinement argument where, as here, there is a related but separate policy for partners, to which employees do not have access. Further, in Walk, the parties conceded, without any substantive analysis by the Third Circuit, that the benefit plan at issue was governed by ERISA as it pertained to both partners and employees. See Wolk v. UNUM Life Ins. of Amer., 186 F.3d 352, 355 (3d Cir.1999). The Third Circuit therefore only addressed whether, as a matter of law, the plaintiff partner, who was designated to receive benefits under *459the employee benefit plan, qualified as a plan “beneficiary.” It was not called upon to decide the issue facing us here. Finally, it is unclear from the facts presented in Lain v. UNUM Life Ins. Co. of Amer., 27 F.Supp.2d 926 (S.D.Tex.1998), to what extent the partner’s plans were related to the employee’s plans. In any event, Lain is not binding on this Court and its summary distinguishing of Robertson should not influence our analysis.
In sum, the parties have cited no authority supporting the argument that merely because two policies are related to each other, the rationale of Robertson does not apply. The authority from our Circuit and others supports the opposite conclusion. Respectfully, I would affirm on this issue under either the clearly erroneous standard or the de novo standard of review.
II. Total Disability Definition
I also believe the opinion is in error in reversing the district court’s determination that House was totally disabled under the policy. As AUL concedes, ERISA does not preempt Louisiana law regulating the business of insurance, such as Louisiana Revised Statute section 22:230 and state court decisions limiting and interpreting insurers’ definitions of total disability in disability income policies. In concluding that House does not qualify as totally disabled, the majority applies decisions that are distinguishable and inapposite here: Ellis v. Liberty Assurance Co. of Boston, 394 F.3d 262, 272 (5th Cir.2004) and Provident Life & Accident Ins. Co. v. Sharpless, 364 F.3d 634, 641 (5th Cir.2004).
Ellis is inapposite because, in that case, we found that ERISA preempted state-law unfair claims processing and good faith and fair dealing claims, and we applied the ERISA abuse of discretion standard in affirming the plan administrator’s denial of total disability income benefits. 394 F.3d at 275-76. As already noted, ERISA preemption does not apply here because Section 22:230 directly regulates the business of insurance and substantially affects the risk-pooling agreements between insurers and insureds. Sharpless is inapposite for similar reasons. There, we determined that ERISA preempted the Louisiana statute under which the insured brought her claim, and thus applied contract interpretation principles pursuant to federal common law.
In the present case, Section 22:230, pertinent state court decisions, and the policy definition at issue required the district court to grant House’s total disability claim. Section 22:230 provides:
A. An individual or group disability loss of income policy to provide loss of income protection against total disability may be issued in this state consistent with the definitions and provisions of this Section.
B. Total disability may be defined in relation to the inability of the person to perform duties but shall not be based solely upon an individual’s inability to:
(1) Perform “any occupation whatsoever”, “any occupational duty”, or “any and every duty of his occupation”; or
(2) Engage in any training or rehabilitation program.
C. A general definition of total disability in such a policy shall not be more restrictive than one requiring the individual to be totally disabled from engaging in any employment or occupation for which he is, or becomes, qualified by reason of education, training, or experience and which provides him with substantially the same earning capacity as his former earning capacity prior to the start of the disability.
D. An insurer may specify the requirement of the complete inability of *460the individual to perform all of the substantial and material duties of his regular occupation or words of similar import.
E. An insurer may require care by a physician other than the insured or a member of the insured’s family,
(emphasis added).
The Louisiana Civil Code provides the general principles for the interpretation of laws as follows: “When a law is clear and unambiguous and its application does not lead to absurd consequences, the law shall be applied as written and no further interpretation may be made in search of the intent of the legislature.” La. Civ.Code Ann. art. 9. “When the language of the law is susceptible of different meanings, it must be interpreted as having the meaning that best conforms to the purpose of the law.” La. Civ.Code Ann. art. 10. “The words of a law must be given their generally prevailing meaning. Words of art and technical terms must be given their technical meaning when the law involves a technical matter.” La. Civ.Code Ann. art. 11. “When the words of a law are ambiguous, their meaning must be sought by examining the context in which they occur and the text of the law as a whole.” La. Civ.Code Ann. art. 12. “Laws on the same subject must be interpreted in reference to each other.” La. Civ.Code Ann. art. 13.
To properly interpret the statute and the policy a court must also take into account Louisiana’s long and unique jurisprudential history of construing such terms in disability income statutes as “total disability,” “occupation,” “regular occupation,” and “all of the substantial and material duties of his regular occupation,” in addition to Section 22:230 and its predecessor statutes. As Louisiana legal practitioner-scholars have recognized, this jurisprudence was part of the background upon which the legislature acted in enacting Section 22:230. William Shelby McKenzie & H. Alston Johnson, III, 15 La. Civil Law TReatise § 290 (3d ed.2007).
Under Louisiana cases, total disability has never required proof that an insured is reduced to a state of abject helplessness. Laborde v. Employers Life Ins. Co., 412 So.2d 1301, 1304 (La.1982) (citing Crowe v. Equitable Life Assurance Soc’y of the United States, 179 La. 444, 154 So. 52, 54 (1934); Madison v. Prudential Ins. Co. of America, 190 La. 103, 181 So. 871 (1938); Nomey v. Pacific Mut. Life Ins. Co., 212 La. 820, 33 So.2d 531 (1948); and Pearson v. Prudential Ins. Co. of America, 214 La. 220, 36 So.2d 763 (1948)). In Johnson v. State Farm Mutual Automobile Insurance Company, 342 So.2d 664, 667 (La.1977), the Louisiana Supreme Court relied on this line of cases to interpret an insurance policy defining total disability as the inability to engage in “every duty of his occupation” to mean the inability to “perform the substantial and material part of his occupation in the usual and customary way.”
Accordingly, the Louisiana Supreme Court, in Laborde, announced that, where an insurance policy expressly purported to adopt the interpretation AUL urges here, the policy would instead be construed to define total disability as “whether he could have performed the substantial and material part of his occupation in the usual and customary way.” 412 So.2d at 1304. In fact, the policy definition of total disability in Laborde was virtually identical to the one the majority applies: “complete inability of the Insured ... to perform every duty pertaining to his occupation.” In Johnson, total disability was similarly defined as the inability to perform “every duty of his occupation.” 342 So.2d at 667. Notably, the liberal meaning imposed in place of these nominally strict definitions omitted the word “every.”
*461Subsequent to these cases, the Louisiana Legislature in 1991 enacted Louisiana Revised Statute section 22:230, a provision plainly intended to limit the ability of insurance companies to adopt overly restrictive definitions of total disability. A leading Louisiana insurance law treatise, William Shelby McKenzie & H. Alston Johnson, III, 15 La. Civil Law Treatise § 290 (3d ed.2007), recently explained how the enactment of Section 22:230 had substantially incorporated the Louisiana courts’ jurisprudential total disability standards:
There is no reason to expect that our courts will abandon the liberal interpretation accorded the concept of disability. Indeed, there may be less reason to do so than there would be in the field of workers’ compensation.... In the health and accident field, the cost is often borne directly by the insured, who has a much greater argument that he should be entitled to a fair shake on the coverage that he paid for.
Indeed, recent legislative amendments provide that disability definitions cannot be simply whatever an insurer might want them to be. Act 879 of 1990, now appearing as La. R.S. 22:230, establishes a definition of disability beyond which an insurer apparently may not go. Coverage for total disability obviously may be offered in Louisiana, but with certain restrictions.
Id. (footnotes omitted).
With this background in mind, applying the Louisiana Civil Code principles of legislative interpretation, it is evident that the purpose of Section 22:230 is to protect the interests of the insured by placing definite limitations on how insurers may define and apply “total disability” in disability income policies. Paying careful attention to the terms of art as defined by the Louisiana courts in previous cases, construing the statute’s provisions in reference to each other, in context, and in conformity with the purpose of the law, the statute in essence provides for: (1) Prohibition of defining total disability in the absolute, unqualified sense: Insurers may not define “total disability” solely as inability to perform “any occupation whatsoever”, “any occupational duty”, or “any and every duty of his occupation”; or as inability to “[ejngage in any training or rehabilitation program;” (2) Maximum level of restrictiveness in such definitions: Insurers may define “total disability” in general no more restrictively than as “totally disabled from engaging in any employment or occupation for which [the insured] is, or becomes, qualified by reason of education, training, or experience and which provides him with substantially the same earning capacity as his former earning capacity prior to the start of the disability;” and (3) Specific requirement of insured’s complete inability to perform all duties of insured’s regular occupation: Insurers “may specify the requirement of the complete inability of the individual to perform all of the substantial and material duties of his regular occupation or words of similar import.” Thus, insurers may condition total disability benefits on the insured individual’s complete inability to perform all of the important duties of his regular occupation, but to do so requires that the insurer specify this requirement of such complete inability in the policy so that it is called to the insured’s attention and agreed to by him.
The disability loss of income policy in this case defines total disability as follows:
“TOTAL DISABILITY and TOTALLY DISABLED mean that because of Injury or Sickness the Person cannot perform the material and substantial duties of his regular occupation.”
The majority misreads the requirements of Section 22:230 and simply disregards *462the Louisiana jurisprudence defining the terms of art used in the statute. Rather, it mistakenly applies the inapposite decisions in Ellis and Sharpless, and, in my opinion, incorrectly concludes that the policy definition must be construed to mean that “in order to obtain total disability benefits, House would have to demonstrate that he cannot perform all of the material and substantial duties of his occupation .... [TJhis means a claimant is totally disabled only if he or she cannot perform each and every material and substantial duty of his or her occupation.” Moreover, the majority also declares “the district court’s distinction between ‘trial lawyer’ and ‘lawyer’ too fíne under a common sense interpretation of ‘regular occupation.’ ” Thus, according to the majority, under the policy definition, an attorney would have to be unable to perform each and every material and substantial duty of the legal profession in general in order to qualify for total disability benefits.
I respectfully disagree with the majority’s conclusions for several reasons. First, the majority’s definition of total disability as inability to perform every duty of lawyers generically is even more restrictive than the “any and every duty of his occupation” definition prohibited by Section 22:2S0(B)(1). Use of the term “regular occupation,” in Louisiana and in general, means the individual insured’s usual and customary means of earning a livelihood, and does not permit the insurer to define total disability at an unreasonably high level of generality so as to offer the insured no real protection in the event he becomes disabled to perform the duties required by his previous regular income-earning activities. As a lawyer, House was still qualified and able to work as a civil servant or governmental attorney, but as a regular occupation that role did not provide him with the earning capacity of his pre-disability high-stress job as an active, experienced actual trial lawyer.
Second, the Louisiana courts have never allowed “total disability” to be defined or applied so restrictively against insured policyholders’ interests. See Laborde, 412 So.2d at 1304 (rejecting an insurer’s advocacy of a definition of total disability similar to that of the proposed opinion in favor of reading the test as “whether he could have performed the substantial and material part of his occupation in the usual and customary way”); Johnson, 342 So.2d at 667 (same).
Third, the Louisiana courts’ interpretation of total disability policy definitions, as well as that adopted by the statute, is in accord with that of state courts’ in general. See, e.g., 46 C.J.S. Insurance § 1089:
Total disability does not mean absolute helplessness, or inability to do anything, or inability to engage in any kind of business pertaining to insured’s occupation, or absolute lack of earning power. However, an inability to perform some aspects of insured’s occupation does not rise to the level of total disability.
Total disability exists when one is wholly disabled from pursuing the usual and customary duties of his employment on which he must depend for a living, or any substantial part of his ordinary duties, or when the injury is such that common care and prudence require insured to desist, and he does in fact desist, from transacting his business.
Generally, when an insurance contract refers to an occupation in promising payments if insured is unable to work, the occupation referred to is the occupation that insured was carrying on at the time that he was injured. When insured lists an occupational description on his application that varies from what he ac*463tually does, the court will look to his actual duties, not the application, to determine insured’s occupation.
(footnotes omitted) (collecting and citing authorities).
Fourth, as interpreted and applied by the majority, the policy definition would be more restrictive than the maximum level permitted by Section 22:230(C). That is, it would disqualify House for total disability benefits even though in his present law practice capability he does not have substantially the same earning capacity as his former earning capacity prior to the start of the disability; thus, the majority imper-missibly interprets “total disability” more restrictively than permitted by law.
Fifth, AUL in its policy did not take advantage of the opportunity offered by Section 22:230(D) for it to “specify the requirement of complete inability” of House “to perform all of the substantial and material duties of his regular occupation or words of similar import.” The policy merely uses a boiler-plate: “TOTAL DISABILITY ... mean[s] that because of Injury or Sickness the Person cannot perform the material and substantial duties of his regular occupation.” Thus, because AUL did not obtain House’s agreement to the specific requirement that, to recover total disability benefits, House must have complete inability to perform all of the substantial and material duties of his regular occupation, as authorized by Louisiana Revised Statute 22:230(D), neither AUL nor this Court can define “total disability” contrary to the requirements of Section 22:230(A)-(C).
III. Conclusion
For these reasons, I respectfully submit that the Court should affirm the district court’s judgment because that court did not commit: (1) a clear error of fact or an error of law in finding that the partner-employer’s disability income policy was separate from the law firm’s ERISA-pro-tected and governed employees’ group insurance; or (2) an error of law in interpreting and applying state law to the total disability definition in the disability income policy of House, the partner-employer.