Clifford C. Coleman, Plaintiff-appellant, v. Interco Incorporated Divisions' Plans, Defendant-appellee, 933 F.2d 550 (7th Cir. 1991)Annotate this Case
Kevin J. Richter, Mathis, Marifian & Richter, Belleville, Ill., for plaintiff-appellant.
James K. Pendleton, St. Louis, Mo., for defendant-appellee.
Before POSNER, FLAUM, and RIPPLE, Circuit Judges.
POSNER, Circuit Judge.
This is a suit under ERISA (Employee Retirement Income Security Act, 29 U.S.C. §§ 1001 et seq.) for retirement benefits. The plaintiff, Clifford Coleman, now 74 years old, was first hired by Interco (formerly International Shoe Company) in 1935. He worked there until 1965 when he left to work for another company. He returned to Interco on July 29, 1969, and became a participant in the company's ERISA retirement plan when it was formed in 1976. In April 1979, only a few months before the tenth anniversary of his reemployment by Interco, Coleman quit because he didn't feel equal to the physical demands of his job as a foreman in a factory; it seems he had difficulty both with walking and with hearing. Had he continued in the job until the anniversary, his pension rights, all agree, would have vested; and no doubt he could have achieved vested status without endangering his health if, instead of quitting, he had taken a medical leave of absence. At argument his lawyer told us that Coleman had not been advised of this possibility, but there is no claim that the company is estopped to enforce the terms of its retirement plan by the advice given or not given to Coleman at the time of his resignation. The company claims that the plan required Coleman to complete ten years on the company's payroll, and that it was a lawful requirement under ERISA. The district judge agreed with both points, and therefore granted summary judgment for the company and dismissed the suit.
The plan is no masterpiece of lucidity, but it makes clear that the employee must, for his entitlement to a pension to vest, have ten years in the company's service. These "Vesting Years-of-Service," as the plan calls them, are defined as "all years of Service (and fractions thereof) measured from the latest date of the Participant's employment by the Employer." "Service" is defined as "all periods of an Employee's employment with the Employer," excluding, "with respect to periods of employment prior to the Plan Effective Date, ... all periods of employment prior to the latest date of employment." From this second provision it is apparent that "latest date of employment" means the most recent date on which the employee was reemployed following a break in service. So you start with July 29, 1969, and count the years and fractions of years until Coleman quit sometime in April 1979 (oddly, the parties cannot agree on the day)--and it doesn't add up to ten years. The time he worked for the company before he left in 1965 doesn't count. For that was before the plan became effective, and the plan expressly excludes such time where, as in Coleman's case, it preceded a break in service that itself occurred before the plan went into effect. This is a lawful exclusion. 29 U.S.C. § 1053(b) (1) (F); Redmond v. Burlington Northern Railroad Company Pension Plan, 821 F.2d 461, 466-68 (8th Cir. 1987).
All this seems clear enough, but Coleman argues that the plan actually is ambiguous because ERISA, though not the plan, defines "year of service" to mean "a calendar year, plan year, or other 12-consecutive month period designated by the plan ... during which the participant has completed 1,000 hours of service." 29 U.S.C. § 1053(b) (2) (A). In the period of twelve consecutive months that began on July 29, 1978, the ninth anniversary of Coleman's reemployment, he worked at least a thousand hours and doubtless more--for he was a full-time worker, and was not laid off at any time during the more than eight months that he worked--albeit he was no longer on the payroll when the twelve-month period ended. Interco's plan uses the statutory term, "year of service," and presumably--Coleman argues--with the same meaning as in the statute. Not so; it is apparent that the plan's term does not bear the statutory meaning, if that meaning is as Coleman argues. The reference in the plan to fractions of years is the essential clue. The only purpose of counting fractions is to see whether the total is ten years, but under Coleman's submission this doesn't matter. The total could be nine years. Suppose he worked a thousand hours in 1969, though he was not hired till the end of July; worked full-time from 1970 through 1977; and worked a thousand hours in the first half of 1978, and then quit. By his computation, he would have completed ten years of service.
Coleman's main argument is that since ERISA requires the vesting of pension rights in a plan subject (as this plan is acknowledged to be) to ERISA after ten years of service, it is the statutory definition rather than the definition in the plan that determines his right to a pension. He is certainly correct that if the definitions diverge, ERISA is trumps. But a Treasury Regulation implementing ERISA expressly authorizes the "elapsed time" method of computing years of service. 26 C.F.R. Sec. 1.410(a)-7(a). (ERISA delegates overlapping rulemaking authority to Labor and Treasury with respect to minimum vesting periods, but an interdepartmental allocation of responsibility has placed regulatory authority concerning years of service in Treasury. 43 Fed.Reg. 47713 (Oct. 17, 1978).) That is the method used in Interco's plan, as in most retirement plans. A year of service in an elapsed-time plan is a twelve-month period, or a bunch of shorter periods tacked together to add up to twelve months, in which the employee was on the company's payroll.
So if the regulation is valid (as held in Swaida v. IBM Retirement Plan, 728 F.2d 159 (2d Cir. 1984) (per curiam)), Coleman loses. He argues that the statute is clear in his favor, in which event, since the Treasury Department has not been authorized to revise the statute, the regulation must be invalid. But the statutory definition of year of service could mean simply that the employee must work at least a thousand hours in a year for the company to be required to count the year as a full year toward the employee's achieving vested status; it is not enough that he was on the payroll for the full twelve months, if his aggregate working time during this period did not reach a thousand hours. This interpretation, which is devastating to Coleman's claim, finds support in another provision of ERISA: "In the case of any seasonal industry where the customary period of employment is less than 1,000 hours during a calendar year, the term 'year of service' shall be such period as determined under regulations of the Secretary." 29 U.S.C. § 1053(b) (2) (C). We are given some pause, however, by reflecting on the plight of a seasonal worker who works more than a thousand hours a year but is taken off the payroll when he is not working, so that under the elapsed-time method he never completes a single year of service, let alone ten such years. (Coleman makes nothing of this point, however.) The provision of ERISA dealing not with when plan benefits vest, but with who is entitled to participate in the plan, excludes an employee who "was separated from the service before the date" specified as the last date on which an eligible employee is entitled to begin his participation. This might appear to exclude the seasonal worker, yet the legislative history indicates that seasonal layoffs are not to be considered separations from service. 29 U.S.C. § 1052(a) (4); H.R.Rep. No. 1280, 93d Cong., 2d Sess. 262-63 (1974), U.S. Code Cong. & Admin. News 1974, 4639, 5038. Further complicating the picture, the Treasury Regulation requires an employer who is using the elapsed-time method to stitch together periods of interrupted service in determining whether benefits have vested. 26 C.F.R. Sec. 1.410(a)-7(d) (1) (ii). This would take care of the seasonal worker who worked more than 1,000 hours a year. But the regulation seems not to apply to seasonal workers at all. Id., Sec. 1.410(a)-7(d) (3). Although we are left with a puzzle about the treatment of seasonal workers, we do not think it unravels the entire regulation. And this is not an interrupted-service case, since we must disregard the period of service that ended before the plan was adopted. Coleman worked consecutively; he just didn't work for ten years.
The elapsed-time method of determining pension entitlements was the most common method used at the time ERISA was passed, Swaida v. IBM Retirement Plan, 570 F. Supp. 482, 489 (S.D.N.Y. 1983), aff'd, 728 F.2d 159 (2d Cir. 1984); it is the most common still. It has the great virtue of certainty and simplicity. The employer doesn't have to keep a record of the hours worked by his full-time employees in order to know when they are entitled to retire with a pension, as he would have to do under the method urged by Coleman, for whom 1,000 hours equals 1 year of service. For then in, say, the fifth month of the tenth year of his service an employee might come to his employer and say, "Surprise. I've just completed my one thousandth hour this year (I've been counting), and now I'm retiring. Bye."
ERISA was a carefully considered statute, and if its framers had intended to wipe out the elapsed-time method of computing pension entitlements we think they would have chosen a more conspicuous method than the obscure wording of the definitional provision on which Coleman relies. Inferences from silence are perilous; but in the event that the hypothesized intention had been expressed, the legislative history would probably contain a record of protest from the business community against such a disruption of established and economical methods of doing business.
The elapsed-time method is valid, and was the method in Interco's plan. It is unfortunate that Mr. Coleman was not better advised back in 1979 about his options as he neared the point of vesting but found himself physically unable to continue working. But his pension entitlement did not vest, and the judgment dismissing his suit must therefore be