One Cecil Sears, now deceased, was an employee of the Internal Revenue Service of the United States. As a federal employee he was covered by a life insurance policy made available to him through the Federal Employees' Group Life Insurance Act, 5 U.S.C.A. § 2091 et seq. The question presented in this case is to whom shall the proceeds of this policy be paid.

Robert Sears and LaVonne Stern, appellants herein, are the children of the deceased, and they claim that they are entitled to the life insurance proceeds because they contend that no valid designation of beneficiary was made in accordance with the provisions of the policy.1 Karen Austin, appellee herein, claims the proceeds by virtue of a holographic will, executed by the deceased on January 20, 1958, which reads in part as follows:

"In all due respect for my son and adopted daughter, Robert Cecil and LaVonne I must remember the time and care given to me by Karen when I was very ill and had no one to help me except her.

"For this I am requesting that all my personal belongings in my apartment — my banking account — my insurance policy with the Federal Government or any other asset be given to Mrs Karen Austin — I hereby give $1.00 to each and any other who may claim any right to my estate."

The proceeds of the policy have been deposited by the Metropolitan Life Insurance Company into the registry of the court, and the insurance company is no longer a party to the litigation.

The district court found that the proceeds of the policy should be paid to Karen Austin, and appellants have filed a timely appeal from that decision. Jurisdiction in the district court was based on 28 U.S.C.A. § 1332, and this court has jurisdiction of the appeal under the provisions of 28 U.S.C.A. § 1291.

The decedent retired on October 31, 1956, without having made a written designation of beneficiary as provided in the policy. As is usual with federal employees, he had not been furnished with a copy of the policy itself but had been furnished with a Federal Employees' Group Life Insurance Retired Employee's Certificate which consisted of one printed sheet. This certificate explains in general terms the rights and benefits to which the holder is entitled. Upon the reverse side of the certificate under the heading "Who Receives Your Insurance Benefits?" the following is printed:

"You do not need to name a beneficiary if you are satisfied to have your life insurance benefits paid in the order of precedence noted below. If you are survived by a designated beneficiary, the benefits will be paid to the beneficiary. If there is no designated beneficiary surviving, the benefits will be paid to your widow or widower under category (1); and if you have no survivor falling in category (1), the benefits will be paid to the survivors falling in category (2); and so on, as necessary, to the other categories.

(1) Your widow or widower.

(2) Your child or children in equal shares, with the share of any deceased child distributed among the descendants of that child.

(3) Your parents in equal shares or the entire amount to the surviving parent.

(4) The duly appointed executor or administrator of your estate.

(5) Your next of kin under the laws of your State of domicile at the time of your death.

"If you wish to name a person or persons not included in the categories above, or to name a person or persons listed but in a different order, you should designate a beneficiary. You may secure the proper form to name a beneficiary or to change the designation from the U. S. Civil Service Commission, Washington 25, D. C. To be valid, your designation or change of beneficiary must be in writing, signed and witnessed, and must be received by the Commission before your death. A witness to the designation may not receive payment as a beneficiary. You do not need the consent of anyone to change your beneficiary. You do not need to fill out a new form or notify the Commission when your address or that of a beneficiary changes."

Portions of 5 U.S.C.A. § 2093, relating to the payment of claims under group life insurance policies, and portions of Section 11 of Cecil Sears' policy are set forth in a footnote.2 These provisions establish the procedure to be used to designate a beneficiary or to change the designation of a beneficiary.

Cecil Sears died on August 8, 1958, and his will was probated in the California courts. Appellants have made no objection to the validity of the will, nor did they file any will contest of any kind. The sole question on this appeal is whether the designation of Karen Austin made by Cecil Sears in his will is sufficient to entitle her to the proceeds of the policy.

The district court found that the decedent intended that the appellee have the proceeds of the policy and made the following statement:

"Unquestionably, decedent failed to follow the prescribed procedure in executing the designation of plaintiff, but it is equally free of doubt that the will manifested an unequivocal intention that the proceeds of the policy should accrue to her. Considering the particular circumstances of his case, and in view of the fact that decedent was a sick and ailing man, the execution of the will can be deemed a reasonable effort on his behalf to execute the designation of plaintiff as beneficiary." [180 F. Supp. 489.]

It is the correctness of this decision that is attacked on this appeal. The appellants argue that the designation is not effective, regardless of intent, if the provisions of the policy are not complied with. However, we feel that the decision of the district judge is correct, and that Cecil Sears' holographic will did effectively designate Karen Austin the beneficiary of the life insurance policy.

Our attention has not been called to any Court of Appeals decision construing the provisions of 5 U.S.C.A. § 2093, a relatively new statute. A district court decision, Smith v. Metropolitan Life Insurance Co., D.C.N.D.Cal. 1956, 142 F. Supp. 320, 322, held that the controlling factor is the intent of the insured, not compliance with formal requirements.

"The universal rule * * * is that the policy provisions specifying the method of changing the designated beneficiary are for the benefit of the insurer and not the insured, and that the clear intention of the insured should be given effect even though the method specified in the policy is not followed."3

Courts of Appeal have many times passed on situations similar to that presented by the case at bar where National Life Insurance policies were involved. Attempts to change a beneficiary have repeatedly been held effective even though the technical requirements set forth in the policies, and the statutes setting up the insurance plan, have not been complied with.

In United States v. Pahmer, 2 Cir., 1956, 238 F.2d 431, 433, certiorari denied, 1957, 352 U.S. 1026, 77 S. Ct. 592, 1 L. Ed. 2d 597, the court, discussing a purported change of beneficiary in a National Life Insurance policy, said:

"Concededly, in the case before us the substitution of the mother as the insured's beneficiary did not fully comply with these regulations. But in the field of National Life Insurance the cases are legion which hold that in judging of the efficacy of an attempted change of beneficiary `the courts brush aside all legal technicalities in order to effectuate the manifest intention of the insured.' Roberts v. United States, 4 Cir., 157 F.2d 906, 909, certiorari denied 330 U.S. 829, 67 S. Ct. 870, 91 L. Ed. 1278. This case may be cited as typical of this long, unbroken, line of authority."4

We feel that the provisions of this policy setting up the method by which a beneficiary may be designated or changed are for the protection of the insurer, and we do not feel that the technical provisions are placed in the policy to protect the insured against hasty or impetuous action. In the case now before this court, the insurer is no longer a party, and the battle is between possible beneficiaries. Since this is the case, there is no reason to invoke technical provisions designed to protect an insurer against the possibility of double payment. We feel that the clearly manifested intent of the insured should control, and we feel that the federal cases clearly support this position.

Appellant's reliance on Cook v. Cook, 1941, 17 Cal. 2d 639, 111 P.2d 322, is misplaced. In the first place this court is called upon in this case to construe a policy the provisions of which are established by federal law, and federal law not state law is controlling in determining to what extent literal compliance with the policy provisions is necessary in order to effectuate a designation of beneficiary. Cook involved a private contract of insurance, not one written as part of a program established by federal law.

The appellants are in error when they assert that Erie R. R. v. Tompkins, 1938, 304 U.S. 64, 58 S. Ct. 817, 82 L. Ed. 1188, directs this court to apply California law to the case at bar. The insurance policy in question was made available to the federal employee by federal legislation. The decision of this case depends on the construction to be given that legislation. This is a question of federal law not of state law. In the situation here presented it would be unwise to assume that Congress, which has the power to say what law is to govern, intended that the federal courts look to the state law to determine what acts are necessary to change a beneficiary.

The appellants have cited some cases where the federal courts have looked to state law to determine the meaning of the word "child" as used in a federal statute. Grove v. Metropolitan Life Ins. Co., 4 Cir., 1959, 271 F.2d 918; Brantley v. Skeens, 1959, 105 U.S.App. D.C. 246, 266 F.2d 447. The reason for resorting to state law in some instances where Congress has not explicitly developed an independent body of federal law is succinctly set forth in DeSylva v. Ballentine, 1956, 351 U.S. 570, 580, 76 S. Ct. 974, 980, 100 L. Ed. 1415:

"We come, then, to the question of whether an illegitimate child is included within the term `children' as used [in the federal statute]. The scope of a federal right is of course a federal question, but that does not mean that its content is not to be determined by state law, rather than federal law. * * * This is especially true where a statute deals with a familial relationship; there is no federal law of domestic relations, which is primarily a matter of state concern." [Citations omitted.]

It is reasonable to look to state law in a situation where there is a well developed body of state law and no federal law, as is the situation with regard to domestic relations; and the courts have consistently found that this was what Congress intended. However, an entirely different matter is presented by the case now before this court. The construction of the federal statute is, in the first instance, a federal question. There is no well developed body of state law on the issue here presented, nor is there any overwhelming state interest that would suggest that here Congress intended that the courts look to the various states for the procedural rules to be applied in naming beneficiaries in federal insurance policies. In the interest of uniformity it is much better that the federal courts here develop a separate body of law rather than look to state law — and fifty different interpretations. See also Reconstruction Finance Corp. v. Beaver County, 1946, 328 U.S. 204, 66 S. Ct. 992, 90 L. Ed. 1172; United States v. Cambridge Loan & Building Co., 1928, 278 U.S. 55, 49 S. Ct. 39, 73 L. Ed. 180; LaBove v. Metropolitan Life Ins. Co., 3 Cir., 1959, 264 F.2d 233; Tatum v. Tatum, 9 Cir., 1957, 241 F.2d 401.

In the second place, the facts of the Cook case, as outlined below, did not make a very compelling situation for allowing a waiver of the policy provisions.

When the policy in Cook was originally issued the named beneficiary was Horatio Nelson Cook, son of the insured. The insured subsequently remarried, having been divorced from his first wife, and on September 21, 1921, changed the beneficiary, making the new wife beneficiary. In making this change all of the provisions of the policy were fully complied with. The deceased later provided in his will:

"Because my beloved daughter Pauline is well provided for by my former wife who recently inherited from her mother I am leaving forty percent of what I possess to my son Nelson knowing that he will pay his sister the forty dollars a month referred to or more if he is able. A policy of insurance Equitable Life Assurance Co. No. 2,536,262 payable on its face to Horatio Nelson Cook is to be collected by him and paid to her * * *."5

The California Supreme Court held that the provision of the will was not effective to change the beneficiary to Horatio, as trustee. In so deciding the court categorically stated that "no expression in the insured's will purporting to assign his life insurance policy or change the beneficiary can be effective." However, the court also mentioned the fact that the insured had once before changed the beneficiary and had at that time complied with the terms of the policy. The court also recognized that there are occasions when an insured may change beneficiaries without compliance with the strict terms of the policy.

"The rigor of the rule requiring that the method provided by the policy for a change of beneficiary must be followed, is relaxed when the controversy is between conflicting claimants and the insurer is not raising the issue. Pimentel v. Conselho Supremo, etc., 6 Cal. 2d 182, 57 P.2d 131, 133. However, as a general principle, the method prescribed by the policy must be followed, and if it is not, no change is accomplished, unless whatever occurred in that respect comes within one or more of the three exceptions to the rule. Those exceptions are based on equitable principles and are designed to assist an insured in carrying out his intent to change a beneficiary in instances such as when he has done all that he could have done to make the change, there is a waiver, or he was prevented from making the change, and where there was not a literal compliance with the terms of the policy."

Thus the court in Cook was faced with a situation where the deceased had previously made a change of beneficiary complying completely with the terms of the policy and where the will that purportedly made the change erroneously referred to Horatio as the present owner. Even if the California courts would still hold, if presented with a private insurance situation similar to this case, that a beneficiary cannot be changed by will, we are not bound in this case by California law; and we choose to follow the more liberal provisions of the federal law as outlined by the cases cited above.

In the case now before this court the deceased had made no previous designation of beneficiary, and the finding of the trial judge that the deceased intended to make Karen Austin the beneficiary of the policy is amply supported by the evidence. The fact that he did not make his original designation in the exact manner set forth in the policy, should not prevent his definite intention, manifested by the affirmative act of drawing up a will, from being given effect.

The judgment of the district court is affirmed.