Bay Ridge Operating Company v. Aaron Huron Stevedoring Corporation/Dissent Frankfurter

by Felix Frankfurter

United States Supreme Court

334 U.S. 446

Bay Ridge Operating Company  v.  Aaron Huron Stevedoring Corporation

 Argued: Oct. 11, 1948. --- Decided: Jan 12, 1948

Mr. Justice FRANKFURTER, with whom Mr. Justice JACKSON and Mr. Justice BURTON concur, dissenting.

No time is a good time needlessly to sap the principle of collective bargaining or to disturb harmonious and fruitful relations between employers and employees brought about by collective bargaining. The judgment of Congress upon another doctrinaire construction by this Court of the Fair Labor Standards Act ought to admonish against an application of that Act in disregard of industrial realities. Promptly after the Eightieth Congress convened, Congress proceeded to undo the disastrous decisions of this Court in the so-called portal-to-portal cases. Within less than a year of the decision in Anderson v. Mt. Clemens Pottery Co., 328 U.S. 680, 66 S.Ct. 1187, 90 L.Ed. 1515, both Houses, by overwhelming votes that cut across party lines, passed and the President signed, the Portal-to-Portal Act of 1947. What is most pertinent to the immediate problem before us is the fact that because the Fair Labor Standards Act had been 'interpreted judicially in disregard of long-established customs, practices, and contracts between employers and employees,' Congress had to undo such judicial misconstruction because it found that 'voluntary collective bargaining would be interfered with and industrial disputes between employees and employers and between employees and employees would be created.' ^[1] Because the present decision is heedless of a long-standing and socially desirable collective agreement and is calculated to foster disputes in an industry which has been happily at peace for more than thirty years. I deem it necessary to set forth the grounds of my dissent.

The Court's opinion is written quite in the abstract. It treats the words of the Fair Labor Standards Act as though they were parts of a cross-word puzzle. They are, of course, the means by which Congress sought to eliminate specific industrial abuses. The Court deals with these words of Congress as though they were unrelated to the facts of industrial life, particularly the facts pertaining to the longshoremen's industry in New York. The Court's opinion could equally well have been written had the history of that industry up to 1916 not been an anarchic exploitation of the necessities of casual labor for want of a strong union to secure through equality of bargaining power fair terms of employment. See, e.g., Barnes, The Longshoremen (1915), passim. Through such bargaining power the agreement was secured which the Court now upsets. Through this agreement, the rights and duties of the industry-the members of the union on the one hand and the employers on the other hand-were defined, and the interests of the men, the employers, and, not least, the community were to be adjusted in a rational and civilized way. On behalf of a few dissident members of the union, but against the protests of the union and of the employers and of the Government, the Court dislocates this arrangement and it does so by what it conceives to be the compulsions of § 7(a) of the Fair Labor Standards Act. ^[2] This is to attribute destructive potency to two simple English words-'regular rate'-far beyond what they deserve.

Employment of longshoremen has traditionally been precarious because dependent on weather, trade conditions, and other unpredictables. Decasualization of their work has been their prime objective for at least sixty years. They have sought to achieve this result by inducing concentration of work during weekday daytime hours.

One of the strongest influences to this end is to make it economically desirable. And so the union has sought and achieved an addition to the basic-the regular-rate sufficiently high to deter employers from assigning work outside of defined periods, except in emergencies. Since 1916, when the International Longshoremens Association made its first collective agreement with waterfront employers in New York, a 50% premium on night and weekend work has generally prevailed. In the industry, this has been colloquially called 'overtime' pay.

Longshoremen do not usually work continuously for one employer, but shift from one to another, wherever employment can be found. The Fair Labor Standards Act does not entitle an employee who works a total of over forty hours per week for several employers, but not more than forty hours for any one of them, to any overtime pay. In view of the peculiarities of this industry, therefore, the only effective way of promoting the aim of the Fair Labor Standards Act, to deter a long workweek, is that devised by the collective agreement, namely, to limit to approximately the statutory maximum of hours the total length of the periods in the week for which additional pay amounting to overtime rates need not be paid, regardless of the employer for whom the work is done.

During the period (1943-45) in controversy, the wage rates were governed by the 1943 General Cargo Agreement between the International Longshoremens Association and the employers at the Port of New York. Under its terms, the 'basic working week,' for which 'straight time' hourly rates were paid, included the hours of 8 a.m. to noon, and 1 p.m. to 5 p.m., Monday through Friday. and 8 a.m. to noon on Saturday. ^[3] 'Overtime' rates, for 'all other time,' were in almost all instances ^[4] 150% of the 'straight time' rates. The 1943 Agreement embodied the practice of the industry since 1916, whereby approximately 150% of 'straight time' rates was paid for night and weekend work. Through the years, with successive renewals of agreements between the International Longshoremens Association and the employers, the rates of pay have risen and the length of the 'basic working week' has decreased. The respondents, members of the International Longshoremens Association, did a large part of their work for the petitioners outside of the enumerated 'straight time' hours. In accordance with the collective agreement, they received, for whatever work they did during the 'basic working week,' 'straight time' pay, and for work at all other times, 'overtime' pay, drawing such 'overtime' pay regardless of whether such work was or was not part of their first forty hours of work in the week. ^[5] They instituted this action, for double damages under § 16(b) of the Fair Labor Standards Act, 52 Stat. 1060, 1069, 29 U.S.C. § 216(b), 29 U.S.C.A. § 216(b), asserting that night and weekend work had been so frequent an incident of their employment that the contractual 'straight time' pay could not be deemed their 'regular rate' of pay, under § 7(a), but that their 'regular rate' was the average of what they received for all their work for any one employer, 'straight time' and 'overtime' together. On this theory, rejected by the union, the employers, and the Government, but now accepted by the Court, all work beyond forty hours per week for any one employer should have been paid for at one and one-half times this average.

The statutory phrase 'regular rate' is not a technical term. Thirteen expressions used in the Fair Labor Standards Act were defined by Congress in § 3. 'Regular rate' was left undefined. The legislative history of the phrase reveals only that it replaced 'agreed wage' in an earlier draft, but there is no indication that this modification had significance. Nor is there any indication that in the field of labor relations, 'regular rate' was a technical term meaning the arithmetic average of wages in any one week. If ordinary English words are not legislatively defined, they may rightly be used by the parties to whom they are addressed to mean what the parties through long usage have understood them to mean, when the words can bear such meaning without doing violence to English speech. The 'regular rate' can therefore be established by the parties to a labor agreement, provided only that the rate so established truly reflects the nature of the agreement and is not a subterfuge to circumvent the policy of the statute. Walling v. Youngerman-Reynolds Hardwood Co., 325 U.S. 419, 424, 65 S.Ct. 1242, 1244. Thus the problem before us is whether the designation of 'straight time rates' for the 'basic working week' in the longshoremen's collective agreement was an honest reflection of the distinctive conditions of this industry.

We are not concerned with an abstract 'regular rate' of pay, for industry is not. The 'regular rate' in a given industry must be interpreted in the light of the customs and practices of that industry. The distinctive conditions of the longshoremen's trade, where employees frequently work during one week for several different employers, are reflected in the provisions which the industry has made in determining rates of compensation. These provisions were designed to secure for longshoremen protection not only from harmful practices common to many industries and dealt with specifically by the statute, but also from those peculiar to the longshoremen's industry, requiring special treatment.

The respondents' wages, as part of a comprehensive arrangement for the betterment of the longshoremen's trade-also covering health and sanitary provisions, minimum number of men in a gang doing specified types of work, 'shaping time,' minimum hours of employment for those chosen at a 'shape,' arbitration, etc.-were determined by a collective agreement entered into between the union and the employers. The Fair Labor Standards Act was 'intended to aid and not supplant the efforts of American wr kers to improve their own position by self-organization and collective bargaining.' H.Rep.No. 1452, 75th Cong., 1st Sess., p. 9. 'The right of individual or collective employees to bargain with their employers concerning wages and hours is recognized and encouraged by this bill. It is not intended that this law shall invade the right of employer and employee to fix their own contracts of employment, wherever there can be any real, genuine bargaining between them. It is only those low-wage and long-working-hour industrial workers, who are the helpless victims of their own bargaining weakness, that this bill seeks to assist to obtain a minimum wage.' Sen.Rep.No. 884, 75th Cong., 1st Sess., pp. 3-4. ^[6] Such assurances were necessary to allay the traditional hostility of organized labor to legislative wage-fixing. The Court now holds unlawful a collective agreement entered into by a strong union, governing the wide range of the longshoremen's employment relationships, and especially designed to restrict the hours of work and to require the same premium as that given by the statute for work done outside of normal hours but within the statutory limit. The Court substitutes an arrangement rejected both by the union and the employers as inimical to the needs of their industry and subversive of the process of collective bargaining under which the industry has been carried on. But, we are told, these untoward consequences are compelled by a mere reading of what Congress has written.

On the question you ask depends the answer you get. If the problem is conceived of merely as a matter of arithmetic you get an arithmetical answer. If the problem is put in the context of the industry to which it relates, and meaning is derived from an understanding of the problems of the industrial community of which this is just one aspect, a totally different set of considerations must be respected. The defendants derived their rights from the entire agreement and not from a part mutilated by isolation. If the parties had written out with unambiguous explicitness that the extra wage in the scheduled periods is to be deemed a deterrent against work during those periods and is not to be deemed a basis for calculating time and a half after the forty hours, I cannot believe that this Court would say that such an agreement, made in palpable good faith, is outlawed by the Fair Labor Standards Act.

How is compensation for services above the limits set by the Act to be reckoned? The standard for compensation could be determined (1) by specific statutory terms; (2) by collective agreement; or (3) by judicial construction in default of either.

Congress could have laid down a hard and fast rule, could have expressed a purely arithmetic formula. It could have said that the rate on which time and a half is to be reckoned is to be found by dividing the total wage by the hours worked. It would not even have been necessary to spell all this out. Congress could have conveyed its thought by using the phrase 'average' instead of 'regular.' And where we have nothing else to go on, except the total wage and the hours, it is reasonable enough thus to ascertain the regular rate. But when parties to a complicated industrial agreement, with full understanding of details not peculiarly within the competence of judges, indicate what the regular rt e is for purposes of contingencies and adjustments satisfied otherwise than by a purely arithmetic determination of the rate of wages, nothing in the history of the law or its language precludes such desirable consensual arrangements, provided, of course, that the parties deal at arms length, and that the defined 'regular' rate is not an artifice for circumventing the plain commands of the law. Such an artifice would obviously not be used in a contract made by workers in their own interests represented by a union strong enough to pursue those interests. Regularity in this context implies of course a controlling norm for determining wages which, though agreed upon between the parties, is consistent with, and not hostile to, the underlying aims of the overtime provision of the Fair Labor Standards Act. Discouragement of overwork and of underemployment are the aims. The longshoremen's collective agreement serves the same purpose as does the statute.

The Fair Labor Standards Act is not a legislative code for the government of industry. It sets a few minimum standards, leaving the main features in the employment relation for voluntary arrangement between the parties. Where strong unions exist relatively little of the employment relation was to be enforced by law. Most of it was left to be regulated by free choice and usage as expressed and understood by the unions and employers. Congress did not provide for increase in basic rates except to the limited extent of establishing minimum wages. The inclusion of such minimum wages is in itself a recognition by Congress of the distinction between what it sought to change and what it sought to use only as the basis for the computation of an overtime percentage.

The claim of the few members in opposition to the union is predicated upon an amount superadded for reasons peculiar to the stevedoring industry to the wage which the parties to the agreement in perfect good faith established as the regular rate. The union members secured this extra wage as part of the entire scheme of the collective agreement. ^[7] This premium is not to be detached from the scheme as though it were a rate fixed by law as a basis for calculating the statute's narrowly limited overtime provision. So long as its minimum wage provisions were complied with, the statute did not seek to change the true basic or 'regular' rate of pay in any industry, from which rate all statutory overtime is to be computed. There is no justification for interpreting the statutory term as including elements clearly understood in the industry to be as foreign to the 'regular rate' as any strictly overtime rates. Here the extra wage is the industry's overtime rate for work which might not be within the overtime period of the Fair Labor Standards Act, but was within the schedule of the collective agreement for extra wages, not because the work was overtime in the ordinary industrial sense but because it was at periods during which all work was sought to be discouraged by making it costly. Because the union secured for its men an extra wage even for not more than forty working hours, the scope of the Fair Labor Standards Act as to overtime is not enlarged. Only for a work-week longer than forty hours is an employee to be paid one and a half times 'the regular rate,' and nothing in the Act precludes agreement between the parties as to what the regular rate should be, provided such agreement is reached in good faith and as a fair bargain. The presupposition of the Act was that voluntary arrangements through collective bargaining should cover an area much wider, and economically more advantageous, than the minimum standards fixed by the Act. The traditional process of collective bargaining was not to be disturbed where it existed. It was to be extended by advancing the economic position of workers in non-unionized industries and in industries where unions were weak, by furthering equality in bargaining power It certainly was not the purpose of the Act to permit te weakening of a strong union by eviscerating judicial construction of the terms of a collective agreement contrary to the meaning under which the industry had long been operating and for which the union is earnestly contending.

There can be no quarrel with the generality that merely because the conditions of employment are arrived at through collective bargaining an arrangement which violates the statute need not be upheld. But this does not mean that in determining whether the contractual designation of certain hours as 'basic' is honest and fair, we cannot consider the fact that the contract was one entered into by a powerful union, familiar with the needs of its members and the peculiar conditions of the industry, and fully equipped to safeguard its membership. To view such a contract with a hostile eye is scarcely to carry out the purpose of Congress in enacting the Fair Labor Standards Act.

The Court has sustained the power of 'employer and employee * * * to establish (the) regular rate at any point and in any manner they see fit,' Walling v. Youngerman-Reynolds Hardwood Co., 325 U.S. 419, 424, 65 S.Ct. 1242, 1245, provided that the regular rate is not computed 'in a wholly unrealistic and artificial manner so as to negate the statutory purposes.' Walling v. Helmerich & Payne, 323 U.S. 37, 42, 65 S.Ct. 11. If we were confronted with an agreement which did not reflect the true practice in the industry, if despite the designation of certain hours as 'basic' and others as 'overtime,' the distinction was not actually observed, but work was done at all times indiscriminately, so that what the contract designated as 'overtime' pay was in reality a 'shift differential,' designed to induce employees to work at less pleasant hours, rather than to deter employers from carrying on at such hours, the labels attached by the parties to the various periods of work would not be allowed to conceal the true facts. We have again and again pierced through such deceptive forms. See, e.g., Walling v. Helmerich & Payne, 323 U.S. 37, 65 S.Ct. 11, 89 L.Ed. 29; Walling v. Youngerman-Reynolds Hardwood Co., 325 U.S. 419, 65 S.Ct. 1242, 89 L.Ed. 1705; Walling v. Harnischfeger Corp., 325 U.S. 427, 65 S.Ct. 1246, 89 L.Ed. 1711. 149 Madison Ave. Corp. v. Asselta, 331 U.S. 199, 67 S.Ct. 1178, 91 L.Ed. 1432, 169 A.L.R. 1293. But here there is no suggestion that the agreement mislabeled the true circumstances of the employment relationship. And it is significant that in no case in which we found that the terms used had distorted the true facts did a union which had made the contract appear to defend it.

The fact that some work was done at odd hours does not misrepresent the regular situation, provided that such work was exceptional and was restricted in frequency by the overtime provisions of the agreement, so that what the agreement treated as regular and what as exceptional were truly just that. We turn then to the actual experience, in representative periods, of the Port of New York longshoremen. The stipulations, exhibits, and findings of the District Court, all demonstrate the exceptional nature of 'overtime' work. ^[8] It is also apparent that such night work as was done was usually done in addition to, rather than instead of, daytime work. The increased compensation for such work therefore served principally to achieve the same result as did the statute namely, to afford a higher rate of compensation for long hours. In peacetime, night work was extremely rate for any one as a recurring experience, and even during the exigencies of war only a small minority was principally so occupied.

The accuracy of the designation of one period or amount of work as 'basic' is not contradicted by the fact that some work may have been done at other times as well. They very reference in any collective agreement to overtime pay for unusual hours implies that some such work is anticipated. A protective tariff need not be so high as to exclude every last item. The statistics in the margin amply justify the trial judge's conclusion that the designations in the collective agreement were not unreal or artificial when the agreement was entered into, and did not become so even at the height of the abnormal wartime effort.

Of course, even if most of the work of longshoremen was performed during 'straight time' hours, if the 50% increment for work at other times was not a true overtime payment, but a shift differential, this higher rate of pay would have to be taken into account in establishing the 'regular rate' of the respondents. But the District Court found that this premium constituted true overtime. As that court stated (Finding 28), a shift differential 'is an amount added to the normal rate of compensation, which is large enough to attract workers to work during what are regarded as less desirable hours of the day, and yet not so large as to inhibit an employer from the use of multiple shifts,' while a true overtime premium 'is an addition to the normal rate of compensation, designed to inhibit or discourage an employer from using his employees beyond a specified number of hours during the week of during certain specified hours of the day. A safe guide for distinguishing between the shift differential and the overtime premium is by the degree of spread between the normal rate and the penalty rate. Whereas a shift differential is usually 5 or 10 cents per hour, the overtime premium is generally 50% of the normal rate.' These findings of the District Court are amply supported by the testimony and by industrial statistics. See 65 Monthly Labor Review 183-85; Wage Structure; Machinery (Bureau of Labor Statistics 1945) p. 21; id. (1946) p. 38; Wage Structure: Foundries (Bureau of Labor Statistics 1945) Tables 32, 33; id. (1946) pp. 44-45. And compare the Directives of the Economic Stabilization Director dated March 8, 1945, and April 24, 1945, limiting the shift differentials which the National War Labor Board could approve to four cents per hour for the second shift and six or eight cents per hour for the third. CCH Labor Law Service, vol. 1A, 10,034.11, 10,462. Applying the test based on Finding 28, and finding also that the differential had in fact served to deter night and week-end work, the District Court held that the fifty per cent increment was true overtime and not a shift differential.

The Court purports to accept the findings of the District Court, and yet it concludes that the District Court erred in finding that the fifty per cent was by way of overtime and not a shift differential. The District Court, to be sure, and not explicitly state that the premium was not a shift differential in one of its formal Findings of Fact. It did so state, however, in its opinion and this conclusion depended on the statements quoted above from Finding 28 as to the characteristics indicativeo f true overtime and shift differentials. I fail to see how this Court can accept Finding 28 and reject the conclusion that the contractual 'overtime' was not a shift differential.

Findings of lower courts are to be disregarded only if not substantiated by the evidence. Here, the evidence supporting the finding was impressive, and yet the Court strains to overturn it to reach a result not urged as socially desirable but only as demanded by legal dialectic. ^[9]

The Court holds that even if the collective agreement accurately designated that regular and overtime work of the generality of longshoremen, it cannot apply to the respondents, because of their particular working hours for a stretch of the wartime period here in controversy. This contention expresses an attitude toward the process of collective bargaining which, if accepted, would undermine its efficacy. It subjects the collective agreement to the hazards of self-serving individualism, which must inevitably weaken the force of such agreements for improving the conditions of labor and forwarding industrial peace. Here, the very increased rates of pay which the respondents received for exceptional right and weekend work was the result of the contract which they now seek to disavow.

Collective bargaining between powerful combinations of employers and employees in an entire industry, each group conscious of what it seeks and having not merely responsibility for its membership but resourceful experience in discharging it, is a form of industrial government whereby self-imposed law supplants force. Cf. Feis, The Settlement of Wage Disputes (1921) c. II. This is an accurate description of the process by which the stevedoring industry has served the greatest port in the United States. Yet the Court rejects the meaning which the parties to the agreement have given it and says it means what the parties reject. Often, too often, industrial strife is engendered by conflicting views between employers and employees as to the meaning of a collective agreement. Here the industry as an entirety-the union and the employers' association-is in complete accord on the meaning of the terms under which the industry has lived for thirty years and under which alone, the parties to the agreement insist, they can continue to live peacefully. But a few members of the union assert an interest different from that of their fellows-some thirty thousand-and urge their private meaning even though this carries potential dislocation to the very agreement to which they appeal for their rights. Unless it be judicially established that union officers do not know their responsibility or have betrayed it, so that what appears to be a contract on behalf of their men is mere pretense in that it does not express the true interests of the union as an entirely, this Court had better let the union speak for its members and represent their welfare, instead of reconstructing, and thereby jeopardizing, arrangements under which the union has lived and thrived and by which it wishes to abide. ^[10]

Collective agreements play too valuable a part in the government of industrial relationships to be cast aside at the whim of a few union members who seek to retain their benefits but wish to disavow what they regard as their burdens. Unless the collective agreement is held to determine the incidents of the employment of the entirety for whom it was secured, it ceases to play its great role as an instrument of industrial democracy. Cf. Rice, Collective Labor Agreements in American Law, 44 Harv.L.Rev. 572; Wolf, The Enforcement of Collective Labor Agreements: A proposal, 5 Law & Contemp.Prob. 273; Hamilton, Collective Bargaining, 3 Encyc.Soc.Sci. 628, 630.

But furthermore, as I read the Court's opinion, it is not limited in application to those employees most or all of whose work was done at night, but extends equally to those work worked chiefly during the 'basic working week,' but also did a few hours of work at other times. Even where a longshoreman worked precisely forty hours of 'straight time,' followed by a few hours of 'overtime' in the same week, payment of the appropriate wages as determined by the collectibe agreement would not satisfy the Court's test that only such extra pay as is given 'for work because of previous work for a specified number of hours in the workday or workweek' ^[11] can be regarded as true overtime pay. To require specification in an industry where the only thing certain is uncertainty is to command the impossible. There is no justification for such a test in the statute, its history, industrial practice, judicial decision, or administrative interpretations. ^[12]

In short, this is not a decision that where the predominant work of an employee is paid for at 'overtime' rates, such rates enter into computation of the 'regular rate,' but rather that where the conditions in an industry are such that the number of 'straight time' hours cannot be precisely predicted in advance, an arrangement for time and a half for all other hours cannot be legal, regardless of how unusual work outside of the 'straight time' hours may be.

But whether or not the Court means to go as fas as it seems to go, and even if its holding is later limited to the narrow situation now before us, I cannot agree with its conclusion. It seems to me that the 'regular rate' of pay for Port of New York longshoremen was the 'straight time' scale provided for by the union contract, and that this was true for the whole union, including the individual respondents. Far from receiving less overtime than the statute required, the respondents were, through the agreement, the recipients of much more. To call their demand one for 'overtime pyramided on overtime' is not to use a clever catchphrase, but to describe fairly the true nature of their claim.

I would reinstate the judgments of the District Court.

Notes

1. Section 1(a), Portal-to-Portal Act of 1947, 61 Stat. 84, 29 U.S.C. § 251(a), 29 U.S.C.A. § 251(a).
2. 'No employer shall * * * employ any of his employees * * * for a workweek longer than forty hours * * * unless such employee receives compensation for his employment in excess of (fory hours) at a rate not less than one and one-half times the regular rate at which he is employed.' 52 Stat. 1060, 1063, 29 U.S.C. § 207(a), 29 U.S.C.A. § 207(a).
3. '2(a) The basic working day shall consist of 8 hours, and te basic working week shall consist of 44 hours. Men shall work any night of the week, or on Sundays, Holidays, or Saturday afternoons, when required. On Saturday night, work shall be performed only to finish a ship for sailing on Sunday, or to handle mail or baggage.
4. For purposes of this case, the 'overtime' rate may be regarded as 150% of 'straight time' in all instances, since the District Court allowed the respondents to recover for those few instances where the 'overtime' was slightly less, and this portion of its judgment was not appealed.
5. On the other hand, although the contract did not so specify, in the unusual situation of a longshoreman working over forty hours of 'straight time' for one employer in one week, he was paid time and a half for the excess. Where this had not been done, the District Court allowed appropriate recovery, and this was not appealed.
6. Similar intentions were expressed again and again in the Committee Hearings and on the floor of both Houses of Congress by the spokesmen of the Administration and Congressional Committee members. See the Joint Hearings before the Senate Committee on Labor and Education and the House Committee on Labor, 75th Cong., 1st Sess., pp. 46-47 (Asst. Atty. Gen. Jackson); id. pp. 181-83 (Secy. Perkins and Sen. Walsh); 81 Cong.Rec. 7650, 7651, 7808 (Sen. Black); 7652, 7799, 7800, 7885-86, 7937 (Sen. Walsh); 7813 (Sen. Pepper); 82 Cong.Rec. 1390 (Rep. Norton); 1395 (Rep. Randolph); 83 Cong.Rec. 7291 (Rep. Allen); 7310 (Rep. Fitzgerald); 9258 (Rep. Randolph).
7. Cf. Lord Stowell, in The Neptune, 1 Hagg.Adm. 227, 232: '* * * the natural and legal parents of wages are the mariner's contract, and the performance of the service covenanted therein; they in fact generate the title to wages.'
8. The following figures were either stipulated by the parties, found as facts by the District Court and concurred in the Circuit Court of Appeals andt his Court, or computed from such statistics:
9. That the hours designated by the agreement as 'overtime' were regarded by the union as excessive hours, rather than merely as unpleasant hours, may also be deduced from the fact that they included much weekday time in which there was ample daylight during a large part of the year, and were not confined to nights and weekends. Another indication of the same thing is the fact that the history of the union agreements for New York longshoremen reveals a succession of reductions of the total number of 'straight time' hours parallel to the reduction of the usual weekly working hours during the same period throughout American industry.
10. Of course, if it can be shown that particular employees for reasons of color, lack of seniority, or anything else-were not fairly or properly represented in the collective bargaining agreement, and were discriminated against and forced into a less desirable class of work, not because of accident or their on desire but because of the deliberate policy of the employers, the union, or both, we cannot treat the agreement made for the generality of longshoremen as binding upon them as well. See Steele v. Louisville & N.R. Co., 323 U.S. 192, 65 S.Ct. 226, 89 L.Ed. 173; Tunstall v. Brotherhood of Locomotive Firemen and Enginemen, 323 U.S. 210, 65 S.Ct. 235, 89 L.Ed. 187. The respondents' claim was not based upon any such allegation.
11. Italics supplied.
12. The Interpretations of the Wage-Hour Administrator pertinent to this case are conflicting and inconclusive. Citation of the most relevant should suffice. Cf. §§ 69, 70, Interpretative Bulletin No. 4, Wage and Hour Division, Department of Labor.