delivered the opinion of the Court.
This case brings here a question as to the application of the overtime provisions of the Fair Labor Standards Act1 to the payment of compensation pursuant to employment contracts similar to those in Walling v. Belo Corp., 316 U. S.624.
Respondent is engaged in the business of cementing, testing and otherwise servicing oil wells, for which it uses its own peculiar equipment. To operate this equipment respondent retains a highly stabilized group of skilled and specially trained “field employees.” The volume of respondent’s business, however, is highly inconstant. *19Consequently, these employees are required to work a variable number of hours from day to day and from week to week.
Prior to passage of the Act in 1938, these employees were paid fixed monthly salaries. Thereafter, they were put on a “weekly-guarantee” plan similar to that which was to be involved in the Belo case. This plan was abandoned March 1, 1942, under pressure from the Administrator of the Act, and reinstated July 1, 1942, after the Belo decision had seemed to end all questions as to its legality.2 Since its reinstatement the plan has been continuously in effect, and embodied in formal written contracts between respondent and the employees to whom it has applied.
The part of these contracts now in issue is respondent’s agreement to pay these employees “a regular basic rate of [a specified number of] cents per hour for the first (40) hours of any workweek, and not less than one and one-half times such basic hourly rate of pay for all time over (40) hours in any workweek, with a guarantee that Employee shall receive for regular time and for such overtime as the necessities of the business may demand a sum not less than $ [a specified number] for each workweek.” 3
The regular basic rate so specified was in each case at or above the minimum prescribed by the Act or by the Administrator’s order, but that rate was always so related to the guaranteed flat sum that the employee became entitled to more than the guarantee only in weeks in which he worked more than 84 hours.4 The compensation *20actually paid was regularly consistent with the contractual obligation as stated.
Petitioner sued respondent under § 17 of the Act to enjoin against future adherence to this plan, on the ground that it failed to include overtime compensation as required by § 7 (a). He contended that the actual “regular rate” of compensation payable under these contracts was not the specified basic rate, but rather the quotient of the amount of the correlative guarantee divided by the number of hours worked in that week. This was said to follow from the fact that the employees usually worked less than 84 hours a week and nevertheless received the full guaranteed sum.
The District Court found, however, that the contracts were “bona fide,” and that they were “intended to and did really fix the regular rate” at which the men were employed. It denied relief5 and the Circuit Court of Appeals affirmed,6 both Courts relying on our decision in the Belo case.
Petitioner admits a close similarity of facts and of his basic contentions in this and the Belo case. He argues, however, that the Belo decision should not be followed: (a) because there are factual differences between the two cases adequate to distinguish them, (b) because Belo has been implicitly overruled by later decisions of this Court, and (c) because the Belo decision is erroneous.
As to the first of these arguments, we note that the contracts in Belo and in this case are substantially identical, except for the amount of the hourly rate and of the fixed guarantee. Under the Belo contract, however, overtime would be paid in addition to the guaranteed wage after 54½ hours had been worked in any given week;7 under *21this contract, only after 84 hours. It is said that this 84 hours bears no relation to the usual workweek.
Actually, the employees in this case have no usual workweek. In many weeks they work more than 100 hours; in others less than 30. In about 20 per cent of the workweeks, they work in excess of 84 hours.8 Whenever they do, they are paid in accordance with the contract on the basis of the specified hourly rate with appropriate overtime.
No more can be said as to the relation between 54½ hours and the usual workweek in Belo. It appears from the record in that case that the employees there involved also worked fluctuating workweeks, and that the average workweek was substantially less than 54½ hours. Indeed, it appears that the Belo employees exceeded 54½ hours in considerably less than 20 per cent of the weeks worked.9 When they did so, they too were paid at the contractual rate with appropriate overtime.10 There is nothing here to suggest different treatment of the two cases.
Petitioner also points to alleged differences in the fact that respondent in this case paid the full weekly guaran*22tee even when its employees worked less than 40 hours in the week, and the fact that respondent carried fixed rather than fluctuating overtime rates on its payroll records.
As to the first of these points, there is actually no difference between this case and Belo. The employees in both cases had a contractual right to the full guarantee however short their workweek, and those in Belo were paid it as well as those here.11 The second fact we think without significance. The function of the payroll records was merely to show the amounts of compensation payable. These records did not affect respondent’s contract obligations, nor suggest a practice at variance with the contract.
We think that whatever differences exist between this ease and Belo are without substance, and that it must either be followed or overruled.
This brings us to petitioner’s second argument, in which our attention is directed to three cases decided since Belo, wherein we held that certain plans of overtime compensation failed to meet the § 7 (a) requirement. It is urged that the provisions for overtime compensation in these cases were legally no less adequate than, and that the principles on which they were decided are necessarily inconsistent with, the overtime provision and the principles of the Belo case.
In Belo itself, the specified basic hourly rate was held to be the actual regular rate because, as to weeks in which employees worked more than 54½ hours, the specified rate determined the amount of compensation actually payable; as to weeks in which they worked less, the Court inferred from the collateral specification of a basic rate and provision for a legal but variable rate of overtime pay that the guaranteed flat sum then due also contemplated *23both basic pay and overtime.12 On the other hand, we find that in the three later cases relied on by petitioner, the agreed method by which wages were computed made a like inference impossible.
In Walling v. Helmerich & Payne, Inc., 323 U. S. 37, we considered a “split-day plan,” under which a prescribed “regular” hourly rate was payable for the first four hours of each eight-hour shift, and a prescribed “overtime” rate, of one and one-half the “regular” rate, was payable for the other four hours.13 In those weeks in which an employee worked statutory overtime, he was paid at the contract “overtime” rate for many straight-time hours and at the contract “regular” rate for many overtime hours. Obviously, these prescribed rates were not actual regular and overtime rates, although so named in the plan. Consequently, as in Overnight Motor Co. v. Missel, 316 U. S. 572, we held that the regular rate was to be determined by dividing the wages actually paid by the hours actually worked. In so deciding, we expressly noted that Belo was not controlling because the wage plans involved in the two cases posed entirely different questions as to the application of § 7 (a).14
*24In Walling v. Youngerman-Reynolds Hardwood Co., 325 U. S. 419, and Walling v. Harnischfeger Corp., 325 U. S. 427, the contracts established two alternative methods for computing each employee’s wages. One was to multiply his straight-time hours of work by a specified basic hourly rate, and his overtime hours by one and one-half that rate, and add the products. The other was to multiply the number of jobs done by a specified piecework rate. The employee was entitled to be paid the greater of these two sums.15 The method of computing the amount due at piecework rates, which were constant for work done on both straight-time and overtime hours, of course negated any possible inference that the payment of such amount contemplated legal overtime compensation. The specified hourly rates were so low, however, relative to piecework rates, that the latter were always, or almost always, determinative of the wage actually to be paid. These cases held merely that such specified hourly rates were not the “regular” rates of wage payments to which they were not related, and which were *25computed according to a necessarily inconsistent method. Again, Belo was expressly distinguished.16
Indeed, it would seem that the Court’s opinions in these cases, far from undermining Beló, showed an affirmative concern that language appropriate to the situations then before us should not be extended to the different situation involved in this and the Belo case.
Finally, petitioner maintains that Belo was wrongly decided and that we should “define the area of [its] continued vitality, if any.” His argument on this score is substantially the same as that advanced on behalf of the Administrator and considered by the Court in the Belo case itself.
The reasons stated in the Belo opinion for rejecting this argument are equally valid today, and need not be repeated. Moreover, our holding in Belo has been a rule of decision in this Court for five years, and recognized as such on each appropriate occasion. Knowing of the Belo decision, the Congress has permitted § 7 (a) to stand unmodified and the courts have applied it as so construed. Employers and employees (including those involved in this case) have regulated their affairs on the faith of it. *26Even if we doubted the wisdom of the Belo decision as an original proposition, we should not be inclined to depart from it at this time.
Affirmed.
52 Stat. 1060, 29 U. S. C. § 201 et seq. The relevant overtime provisions, contained in § 7 (a), 29 U. S. C. § 207 (a), are as follows:
“No employer shall, except as otherwise provided in this section, employ any of his employees who is engaged in commerce or in the production of goods for commerce—
“(3) for a workweek longer than forty hours after the expiration of the second year from such date, unless such employee receives compensation for his employment in excess of the hours above specified at a rate not less than one and one-half times the regular rate at which he is employed.”
This Court decided the Belo case June 8,1942.
Compare the almost identical wording of the Belo contract, 316 U. S. at 628.
For instance, the lowest specified basic rate in May, 1944, when this case was tried, was 40 cents an hour. Compensation at this rate for 40 hours and at one and one-half this rate for 44 additional hours equals $42.40. Actually, the correlative weekly guarantee was the slightly greater sum of $42.69.
57 F. Supp. 408.
152 F. 2d 622.
See the statement and explanation of the Belo contract, 316 U. S. at 627-629.
The record shows that of 4,284 man-weeks worked by respondent’s California field employees between July 5, 1942, and March 11, 1944, about 3% were less than 20 hours in length, about 13% were less than 40 hours, about 67% were from 40 to 84 hours, about 20% were over 84 hours, and about 7% over 104 hours, some running as high as 140 and 150 hours. This “work-week” was the basis for determining compensation, but it did not represent the number of hours actually worked by the employee. In the course of typical cementing and testing operations, many hours counted as working time were spent waiting while the drilling crew was running the casing, the cement was setting, the perforation work was being done by another company, or the testing tool was standing in the well hole.
See Transcript of Record, Supreme Court of the United States, Walling v. A. H. Belo Corp., No. 622, O. T. 1941, pp. 194-337.
316 U. S. at 631-632.
Fleming v. A. H. Belo Corp., 121 F. 2d 207, at 210, note 6.
316 U. S. at 631-632. In Overnight Motor Co. v. Missel, 316 U. S. 572, decided the same day as Belo, the employees also worked an irregular number of hours each week, but were simply paid a fixed weekly wage. The Court noted the absence of any agreement between the contracting parties for the payment of a specified rate and overtime, and of any contractual limitation on the number of hours the employee could be required to work for the fixed wage. It also noted that these factors were not absent from the Belo plan. See 316 U. S. at 581.
Theoretically, when an employee had so accumulated 40 “regular” hours in one week, all subsequent hours were compensable as “overtime.” Actually, no employee ever did so. See 323 U. S. at 41.
“Nothing in this Court’s decision in Walling v. Belo Corp., supra, sanctions the use of the split-day plan. The controversy there cen*24tered about the question whether the regular rate should be computed from the guaranteed weekly wage or whether it should be identical with the hourly rate set forth in the employment contract. There was no question, as here, pertaining to the applicability of the regular rate to the first 40 hours actually and regularly worked, with the overtime rate applying to all hours worked in excess thereof.” Walling v. Helmerich & Payne, Inc., 323 U. S. 37, at 42.
In the Harnischfeger case, the scheme was actually a little different from that in the Young erman-Reynolds case, which is stated in the text. In Harnischfeger, the employee was credited with (a) the product of the total number of hours worked multiplied by the basic rate, plus (b) the amount by which piecework earnings during all hours worked exceeded the product in (a), plus (c) the product of the number of overtime hours worked multiplied by one-half the basic hourly rate. The difference is that in Harnischfeger some provision was made for overtime; but in both cases the provision for overtime was inadequate, and for the same reason. See 325 U. S. at 431-432.
“This Court’s decision in Walling v. Belo Corp., 316 U. S. 624, lends no support to respondent’s position. The particular wage agreements there involved were upheld because it was felt that in fixing a rate of 67 cents an hour the contracts did in fact set the actual regular rate at which the workers were employed. The case is no authority, however, for the proposition that the regular rate may be fixed by contract at a point completely unrelated to the payments actually and normally received each week by the employees.” Walling v. Youngerman-Reynolds Hardwood Co., 325 U. S. 419, at 426.
See also the concurring opinion of Mr. Justice Frankfurter in Walling v. Harnischfeger Corp., 325 U. S. 427, 433. The Court did not expressly refer to Belo in its opinion in the Harnischfeger case; but, as it did in Young erman-Reynolds, which involves substantially the same question and was decided the same day, we consider that further reference to Belo would have been redundant.