Problem Child Among Labor Laws—The Walsh-Healey Act

3ut

law 3ournat

SPRING

PROBLEM CHILD AMONG LABOR LAWS-THE WALSH-HEALEY ACT

RUDOLF MODLEY,* JAMES R. PATRON, JR.t & GERARD D. REiLLYj

I

held full-dress "briefing conferences" on the act.4 It has also established a separate Walsh-Healey Committee and has set up a lawyer. economist team to follow all Walsh-Healey determinations.5

There have also been increasing public references to the adverse effect of determinations on the attainment of national objectives in regard to foreign trade, inflation control, assistance to small business and to depressed areas, etc."

It is thus apparent that concern about the act and its implementation is rapidly growing. Until recently, it was a little-known and even less well understood legislative relic left over from the pre-Fair Labor Standards Act7 days. The explanation for the intensity of present concern about the act can be traced directly to recent changes in the method and timing of its application, changed economic circumstanc es within the United States and this country's increasing competitive problems in international commerce.

Under this act, which has now been on the books for more than a quarter of a century, the Secretary of Labor determines the prevailing minimum wages which government contractors in different industries must pay. Once a determination has been made, contractors entering into a contract for more than $10,000 with the federal government must pay covered employees working on that contract not less than the minimum determined.

A. The Reasons for Industry Concern

Recent industry concern is due primarily to three developments: one, the shift in emphasis in the selection of industries for determination proceedings from low-paying industries to higher-paying ones; second, the announced speed-up in the determination process; and third, the effect of the Secretary's current method of determining industry "minimums," especially for industries in, or emerging In many respects, the method or from, low-employment periods. approach of the Secretary is most subject to criticism-both in principle and in practical effect.8

Many industrial leaders may well agree with the congressional objectives underlying the act which, although not specifically stated in the act itself, have been asserted to be "to eliminate the practice under which the Government is compelled to deal with sweatshops"9 and "to prevent government business going to bidders with wage practices which (are) substandard for the particular industry."'0 However, many industries have become increasingly alarmed by the Secretaries' interpretation of this purpose which has led to determinations which are occasionally higher than minimum wages in force in every company in an industry1' and almost always higher than the minimum wages in force in the majority of all plants.

In conformity with congressional purpose, Secretary Perkins, during whose tenure the law was enacted, focused attention on lowpaying industries where sweatshop conditions were prevalent. In contrast, the Secretary now selects industries for determination proceedings in which there is "the probability of prevailing minimum wages well above the present statutory minimum under the Fair Labor Standards Act."'12 It is obvious that sweatshops are not likely to exist in such industries and that determinations are likely to disturb wage patterns frequently arrived at by collective bargaining and long established employment characteristics.

2. Acceleration of Determination Process

years. The new "procedures" are intended to shorten this period to less than one year."'

The procedural time lag in the past coupled with a continuing upward spiral in wages have helped greatly to soften the adverse impact of determinations. Wage increases between the time for which wage data were available and the time at which wage determinations became effective were often substantial. These increases tended to neutralize the serious inflationary effect due to the curious statistical practices applied by the Secretary in determining a "prevailing minimum." A slow-down in wage increases on the one hand, paired with a speed-up in putting determinations into effect, could aggravate the impact of determinations on many industries and regions.

3. Effect of Low-Employment Periods

The third reason for concern lies in the fact that the statistical practices applied by the Secretary penalize any industry which by hazard is in, or just emerging from, a depression at the time of determination proceedings. This results from the fact that the Secretary elects to ignore the lowest wage rates in an industry's rate structure if they are not being paid at the time of the wage survey14 He interprets the statutory "prevailing minimum wages" as equivalent to the lowest wages which happen to have been paid in a particular week.

This has a highly inflationary and arbitrary effect if the survey is taken at a period of low employment during which workers with low seniority are laid off and therefore not on the payroll. The remaining workers are usually higher paid because of merit or seniority advances. In many plants, there may well be no worker at all in the lowest labor grades, and those that are are likely to be at the higher end of the rate range. The situation is similar in industries just emerging from depressed employment. Added workers are generally drawn from recall lists and do not enter at the lowest

established rates but at rates higher than the minimums in the rate structure.

By basing a determination on these higher wages, the Secretary, in effect, makes it impossible for an industry to return to its true prevailing wage structure when employment conditions return to normal and new and inexperienced workers are hired.15

II

A. History of the Act'

1. The Roosevelt Era

The setting of standards for minimum wages and maximum hours for private industry was a major objective of Franklin Delano Roosevelt's administration. Even before her appointment as Secretary of Labor, Frances Perkins submitted to the President a proposal for such wage and hour legislation. The industry codes set up under the NRA (National Recovery Administration) contained wage and hour standards. When these codes were declared unconstitutional in 1935,17 Miss Perkins proposed legislation which applied to government contractors the same standards of wages, hours, and working conditions which NRA had tried to impose on all industry.'8 While previous laws had provided for an eight-hour day for government contractors, the new legislation covered a much broader field.

Bills were introduced both by Senator Walsh in 1935 and Representative Healey in 1936. Senator Walsh, author of the earlier bill, accepted the Healey version and the Walsh-Healey Act became law on June 30, 1936. The hearings and the House debates19 show that

the Walsh-Healey Act was intended as an interim measure to provide wage and hour standards for government contractors until a broader statute could be enacted which would again cover most of the workers in private industry.

Two Supreme Court decisions in 193720 seemed to clear the way for general -wage and hour legislation. President Roosevelt sent a message to Congress urging such legislation and the Fair Labor Standards Act2' became law in 1938, just two years after the enactment of the Walsh-Healey Act. Continuing doubt as to the constitutionality of the Fair Labor Standards Act may explain why the Walsh-Healey Act was not repealed at the time.

Secretary Perkins' determinations were largely in the field of what were then sweatshop industries-the textile and garment industries. When the statutory minimum under the Fair Labor Standards Act was reached in these industries, Secretary Perkins refrained from making new wage determinations. Moreover, no wage determinations were made between mid-1943 and 1948.

2. Developments Since 1948

In fiscal year 1939, prior to lend-lease and World War II, govIn contrast, ernment contracts accounted for only $528 million. during fiscal year 1961, unclassified contracts alone with a value of almost $17 billion were subject to the Public Contracts Act. Classified contracts probably amounted to a comparable amount. Onethird of this unclassified procurement was with industries in which the Secretary of Labor had made minimum wage determinations under the Walsh-Healey Act.22 The comparatively greater potential in recent years for a larger impact from these determinations on American industry is evident.

Up to March 12, 1962, 55 determinations had established 96 (Eight industries have different rates minimum hourly wage rates. for different industrial segments; two have different rates for different regions.) Of the 96 rates, 40 were at $1.1523 and therefore have no effect on industry pay practices because of the identical minimum 20West Coast Hotel Co. v. Parrish, 300 U.S. 379 (1937); NLRB v. Jones & Laughlin

under the Fair Labor Standards Act. The total distribution of determinations and rates was as follows:

Minimum hourly rate (in cents) 115 116-130 131-150 151-200 201 and over

Wage rates set

40 12 13 10 2125

Between January 1961 and March 1962, 10 determinations or redeterminations with 18 wage rates became effective. Of the 18 wage rates set, 8 were in $1.51 to $2.00 bracket; 7 in the $1.31 to $1.50 bracket, and only 3 were lower, none less than $1.20. The shift to enforced "minimum" wages "well above the present statutory minimum under the Fair Labor Standards Act" has thus effectively been carried out.

In 1948, under Secretary Tobin, new determination procedures were begun. They have continued under subsequent Secretaries to the present date. The period from 1948 to date has also brought about important changes in the statistical techniques used in making determinations, resulting in several instances of proposed or actual regulation which industry has judged unnecessarily burdensome.

B. Interpretation and Implementation of the Act

1. Stipulations Required

The Walsh-Healey Act, and the regulations issued under it, require every government contracting officer to insert in every contract a stipulation by which the person awarded the contract agrees to be bound by the act and the regulations.26 While subcontractors are generally not covered unless found to be "substitute manufacturers," the status of "facility" contracts where goods are furnished

to government-owned but privately-operated facilities, is far from clear.27

In accepting the stipulation, the contractor agrees:

a. that he is either the manufacturer or a regular dealer as de-

fined;

b. that all persons employed in the manufacture or furnishing of the goods under the contract will be paid not less than the prevailing minimum wage determined by the Secretary for this industry;

c. that no employee working on the contract shall work more than 8 hours in any one day or more than 40 hours in any one week except on payment for overtime;

d. that no male person under 16 years of age and no female person under 18 years of age and no convict labor will be employed in performance of the contract; and

e. That no part of the contract will be performed under unsafe

or unsanitary conditions.

This article is primarily concerned with the prevailing minimum

wage standards under the act.

2. Terminology of Section 1(b)

Section 1 (b) says that persons employed will be paid "not less than the minimum wages as determined by the Secretary of Labor to be the prevailing minimum wages ... .'28 The act does not define what is "prevailing," nor does it define what "minimum wages" are. The act places the responsibility of ascertaining what "prevailing minimum wages" are in the hands of the Secretary. Whether this is a wise and judicious repository for such controversial and farreaching powers is not a subject which this article will undertake to assess.29

While it was clearly the intent of the lawmakers that no one in the executive branch have "any authority to fix wages"80 and that "the minimum wages established under the statute ... merely conform to the standards which the industries themselves have generally

adopted,"3' the absence of a clear statutory definition of "prevailing" and of "minimum wages" has enabled Secretaries who are so inclined to make determinations far in excess of what the industries themselves consider as their general minimum wage structure.

3. What Is "Prevailing"

Before a prevailing minimum wage can be found, it must be known "what it is that characterizes wages as 'prevailing,' what distinguishes them from other wages which do not prevail, and what elements not present in other rates must be present in particular rates to impart to them the condition of prevalence. In other words, a definition of prevailing minimum wages is called for."32

As a matter of fact, in most industries there is no wage rate which can be properly considered as "prevailing" in the normal sense of the word use. The typical situation has been described in the proposed decision for the drugs and medicine industry: "In this industry no single minimum hourly wage is found among the several plans with such frequency that it fairly may be said to be 'prevailing' 33 In spite of this, no Secretary has ever refused in the industry .... to make a determination on the grounds that a prevailing wage, in fact, does not exist. However, three completely different techniques have been used by various Secretaries of Labor to arrive at a decision as to what might be called "prevailing": the contract minimum, the cluster technique, and the median technique.

a. The Contract Minimum. There are a few industries in which the contract minimum set in collective bargaining may justifiably be considered as the prevailing minimum. This is the case when collective bargaining is being used by the large majority of plants and employees and the rates agreed upon are identical for the entire nation or for broad geographical areas.

The industries which have minimum wages determined on the basis of the contract minimum are: pressed and blown glass and In these glassware; cement; iron and steel; and bituminous coal. four industries, collective bargaining agreements with a single union covered 80 per cent or more of all employees. Wage rates agreed

upon applied either to all plants in the nation or to all plants or mines in defined geographical areas. The Secretary, however, considers it not sufficient for the minimum contract rates to be agreed upon; they must actually be paid to a substantial number of workers to be accepted by the Secretary as the basis for a determination.

The contract minimum technique has been used consistently since the enactment of the law in the few situations where conditions permitted its use. The major objections against this technique were raised by smaller plants and mines which were not part of the collective bargaining agreements.

b. The Cluster Technique. Where no wage rate could be properly considered as "prevailing," early practice frequently used the "cluster technique" to arrive at a determination. In 1938, the Public Contracts Board gave the clearest description of the principle behind the cluster technique in the Tag Industry case:

A prevailing minimum wage, although lying in the lower part of the wage structure, should therefore be the wage that predominates in that field; it should have superior force and influence in relation to the lower and higher wages which surround it within the lower part of the wage structure.34

In practice, this technique worked as follows: A table of average straight time hourly earnings of all covered employees in the industry was prepared, usually in five-cent intervals. Beginning with the lowest wage interval and moving towards the higher ones, the department searched for the "first significant cluster." This "cluster" showed a larger percentage of employees in that wage interval than were found in the wage intervals immediately below or above it.

Under Secretary Perkins, the Public Contracts Board made an effort to separate the semi-skilled and skilled employees from the unskilled and find the first significant cluster only in an analysis of the wages paid to the unskilled.35 Secretary Tobin applied the technique to tables which included wage data for all covered employees and thus increased the tendency to create clusters made up of wages which were not truly minimum wages.

Objections to this technique were made as follows: The law charges the Secretary with determining the prevailing minimum

wages. To do this, the Secretary should analyze only minimum wages. The minimum wages, by their nature, must be the lowest wages in each plant. The average straight-time hourly earnings tables on which the Secretary based his determinations included not only minimum wages but also the earnings of workers with higher skills receiving higher wages, the earnings of unskilled workers who had advanced beyond the lowest rate because of merit or seniority, and the earnings of incentive workers. The "clusters" found significant by the Secretary were thus possibly "clusters" only because they included wages which could not possibly be called "minimum" wages.

This technique has been abandoned since about 1950, although reference has been made to it on occasion6 in more recent determinations.

c. The Median Technique. Late in 1950, Secretary Tobin heralded the arrival of a new technique in a magazine article.37 He redefined the purpose of the Walsh-Healey Act: "[T]o protect the majority of fairminded employers from being underbid ...and thus to safeguard the working conditions of a majority of the employees."38 In practice, this has placed major emphasis on statistical tables which show the minimum wages paid in the plants of the industry both according to the number of plants and to the covered worker employment in these plants. The "prevailing" minimum is then a figure which is being paid (or exceeded) by approximately half or more of the plants in the industry, employing more than half of all employees covered by the act.

In the tentative decision' for the Electronic Component Industry, the Secretary stated that "the fact that approximately half the industry is in establishments having a lower minimum wage and half in establishments having a higher minimum wage qualifies such medians as best representative of the industry as a whole. ..."1,9 He further confirmed in the same case that "in several instances the Secretary of Labor has found the prevailing minimum wages at some points between the medians of the lowest wages in the several establishments in the industry counting all the establishments as equals, and weighting each for its covered worker employment."40 80 Backman & Levine, The Prevailing Minimum Wage Under the Walsh-Healey

At the present time, the median technique as described is the one most commonly used for industries in which no minimum actually does prevail.40

4. What Are "Minimum Wages"?

The controversy between the Secretary and industry in regard to the meaning of "minimum wages" centers on which criteria should be used: the lowest established rates in force in a plant, or the lowest wage which happens to be paid in that plant at the time of the government survey. Under certain circumstances, the difference, expressed in cents, can be very substantial.

Whenever the making of a prevailing minimum wage determination has necessitated an answer to the question of whether a determination should be grounded upon lowest established rates or the lowest rates actually paid, reliance has been placed by the Secretary consistently upon the lowest rates actually paid.41

Industry generally holds that the Secretary's position is, in many instances, contrary to common sense and results in determinations which are biased against industry, particularly against industries in depressed conditions. When determinations are based on wages actually paid, no consideration may be given to the rate structure actually in force at the plant.

5. Whose "Minimum Wages" Are Studied

a. The Language of the Law. The law states that the Secretary will determine "the prevailing minimum wages for persons employed on similar work or in the particular or similar industries or groups of industries currently operating in the locality ....

,,42

b. The "Particular Industry" Standard. Application of a "similax work" standard might conceivably be interpreted to permit several

wage determinations within one industry on an occupational basis. To date, the Secretary has rejected this interpretation. The fact that there is no standardization of job titles and of job content in most manufacturing industries would make an occupational approach difficult. A member of the early Public Contracts Board found the conclusion "inescapable that 'similar work' as a standard under ..is a clumsy standard, wasteful in its the Public Contracts Act . administrative exigencies anddoubtful in the equity of results that might be produced with its use."43 And he stated that "the Department of Labor has interpreted the language of the act as excluding occupational wages."44

It was therefore a considerable surprise when the Secretary recently proposed (but has not yet implemented) separate determinations for blueprint machine operators and draftsmen, and all other covered workers in the machine tool industry.45 The fact that the Secretary stated in the same proposed determination that he was "seeking to determine a prevailing minimum wage which will be enforced with respect to all employees"48 confirms the belief that the proposal was only an attempt to salvage a proceeding which was deficient because of a faulty definition of covered employees rather than an effort to reverse long standing construction of the statutory language and consistent Secretarial practice.47

The "similar industry" standard has not been used and has been

classified as "at best a superfluous alternative."48

The "groups of industries" standard has significance only because the Secretary has tried to limit the applicability of the locality clause "currently operating in the locality" to this standard only.

In general, the Secretary seeks to determine prevailing minimum wages on the basis of the "particular" industry standard,49 i.e., the minimum wages in the industry to which the determination will be applied.

6. Locality

The ambiguity of the language of section 1 (b) has led to a longstanding controversy. The Secretary has always held that the phrase "currently operating in the locality" qualified only the term "groups of industries." Because he never used "groups of industries" as the basis for a determination, he always felt free to ignore the "locality" language of the act.0

The Secretary, moreover, has taken the position that determinations are to be made on a nationwide or industrywide basis wfiere an analysis of government procurement shows that bids are submitted, and contracts awarded, for delivery on a nationwide basis.51 This interpretation has been upheld by the Court of Appeals for the District of Columbia, and certiorari denied to industry by the Supreme Court. 2

The area for competition is thus presently measured in terms of competition for government contracts only. "Sectional or regional divergencies in minimum wage practices are not necessarily incompatible with industry-wide competition for contracts" according to the Secretary,8 citing the position of the court of appeals in Mitchell 4 that "only an industry-wide minimum will v. Covington Mil serve this (the Walsh-Healey Act's) purpose, because the competition is industry-wide."

The legal wisdom of the refusal of the Supreme Court to review the Secretary's interpretation has since been seriously questioned.55 It seems that the Secretary's position in this regard is at best a strained interpretation of the simple wording of the act; at worst it flies directly in the face of congressional intent as expressed at that time. It is generally thought that the Secretary's position had its origin in, or at least was strengthened by, the attitude of many congressmen from New England whose textile centers were feeling the effects of low-wage competition from the South. It is quite likely that continued nationwide determinations of the Secretary will in the long run bring more economic hardship to New England than to any

other region. What may have appeared at an earlier time to be a possible deterrent to textile manufacturers who considered moving plants to the South, may now prevent New England from being able to compete for government contracts.

7. Open Market

Section 9 of the Walsh-Healey Act excludes from application of the act materials "as may usually be bought in the open market."56 Despite the apparent intent, the Secretary has ruled that the "open market" exemption applies not to products which may usually be bought in the open market, but only to those which the government itself buys in the open market. Whenever the government advertises for bids, it does not buy "in the open market" and the act therefore does apply.57

C. Administration of the Act

Administration of the Public Contracts Act is vested in

the

Department of Labor. A Public Contracts Division was created to administer the Walsh-Healey Act, but in 1942, this division was consolidated with the Wage and Hour Division which had been established pursuant to the provisions of the Fair Labor Standards Act. An administrator, appointed by the President with the consent of the Senate, heads the consolidated Wage and Hour and Public Contracts Divisions.58 Under the administrator are two assistant administrators, one for wage determinations and research, and another for enforcement and program planning.59

determinations must be made on the record after opportunity for a hearing. The first step in a proceeding is generally the convening of a labor-management panel conference to explore definitional and other relevant problems and to work out plans for obtaining the necessary wage data. Subsequently, all interested parties are afforded opportunity to submit evidence on minimum wages and other pertinent issues at a public hearing conducted pursuant to the Administrative Procedure Act. Parties may also file briefs containing proposed findings of fact and conclusions of law. Upon certification by the hearing examiner, the full record of the proceedings is carefully analyzed and the Secretary of Labor issues a proposed decision based on the facts of

1649 Stat. 2039 (1936), 41 U.S.C. §43 (1958). 17 Letter from R. N. Elliot, Acting Comptroller General of the United States, March 24, 1937, in Hearings on S. 2594 and S. 2645 Before the Senate Committee on Banking and Currency, Part 5, 82d Cong., 2d Sess. 2674-75 (1952).

58GSA, UNrrED STATES GOVERNMENT ORGANIZA'rION MANUAL 339-41 (1961-62).

OFFICIAL CONGRESSIONAL DIRECTORY, 87TH CONG., lr Sss. 519 (1961).

record. Opportunity is provided for interested parties to submit exceptions to the Secretary's proposed decision. After reviewing such exceptions, a final decision is issued and the resultant determination becomes effective 30 days after publication in the Federal Register. The determinations ...are applicable only to contracts for which bids are solicited or negotiations otherwise commenced on or after the effective date.0

The procedural description given above by the Department in is still essentially valid. Following industry protests and 1961 union representations, there were a few changes in that procedure under former Secretary Goldberg. In addition to the announced determination of the department to complete wage determinations in one year or less and to accelerate redeterminations, new regulations provided that final decisions in all determinations-even those affecting industries for the first time-were to become effective seven days following publication.61 Industry reaction to this was immediate.62 The basic criticism was that this was insufficient time to allow education of industry members and to permit them to make necessary adjustments in their payroll or to decide whether to continue to do business with the government.

On the plus side, the Secretary partially yielded to the industry contention that those who participated in the hearings on behalf of the Secretary should not also participate in the review and determina"As a guarantee of fairness," comments on tentative tion process. decisions are now said to be reviewed in the Office of the Secretary

"by persons other than those who participated in the hearings or in the drafting of the tentative decisions."0'3 Unfortunately, tentative decisions are still apparently made by participants at the hearing and published under the Secretary's name. This makes any substantial change through review psychologically difficult.

The investigation work of the consolidated divisions is conducted through ten regional offices and one territorial office. Employers and employees subject to the Public Contracts Act are usually subject also to the Fair Labor Standards Act and concurrent investigations are made where both acts applyP

00 48 DEP'T OF LABOR ANN. REP. 249 (1960). 6" Dep't of Labor News Release No. 4774, supra note 13. 03 See, e.g., Brief Supporting Exceptions of the National Machine Tool Builders' Association to the Tentative Decision in the Determination of Prevailing Minimum Wages, Feb. 21, 1962, p. 27.

63 Dep't of Labor News Release No. 4774, supra note 13, at 2. " In the fiscal year 1961, 46 administrative proceedings were instituted under the

In its report to the Bureau of the Budget, the Advisory Council on Federal Reports, Committee on Walsh-Healey Act, proposed that the decision of the Secretary to institute determination or redetermination proceedings in any industry should be subject to review by the Bureau of the Budget and that objective, published criteria on which a selection is made be followed. The Committee also proposed that at the hearings parties should be free to introduce evidence showing that objective criteria had not been followed and

that a determination or redetermination for the industry should not be made.69

As indicated above, the Department has published rather general If criticism is due, it is criteria relating to the selection process. rather that the standards are inadequate and inconsistent with the original objectives of the act. Foreign competition, regional differentials, plant modernization, cyclical conditions, and other social and economic factors are ignored.

A. The Labor-Management Panel Conference

Once a decision has been made to start with proceedings, an informal advisory panel is invited by the Secretary to discuss such items as the proposed definition of the industry, the form of the wage questionnaire to be used in making the wage survey, the payroll period for which questionnaire data are to be reported, the smallest size establishment to be surveyed, lists of the establishments to be included in the survey, and other factors of importance in the deIndustry may be represented on the panel by termination process. its trade association (and legal counsel and economic consultant) as well as by some company representatives. Labor is represented by representatives from AFL-CIO headquarters and by representatives of the labor unions active in the industry. The government is represented by personnel from the Wage and Hour and Public Contracts Divisions, the Office of the Solicitor, and the Bureau of Labor Statistics (BLS). Occasionally, representatives of other government agencies also take part.

The advisory panel performs some truly essential functions in guiding the determination effort. Extensive industry preparation before the panel meeting and selection of the most highly competent technical, legal, and economic industry representatives for the panel In the experience of the activities has been vitally important. authors, industry representatives should be able to speak for establishments of all types and sizes and be familiar with the problems of all regions in which the industry is located. The fact that industry's position-or lack of position-at the preliminary panel conference is frequently quoted in hearings by the representative of the Solicitor's office as purported support for his own position emphasizes the

importance of the most extensive preparation possible by industry representatives on subjects likely to be brought before the panel.70 Many industries, lacking knowledge of the proceedings and the implications of the conference, have often been represented by trade association personnel who were not fully aware of the implications of the items discussed and decided upon. In many instances participation by uninformed industry representatives can be harmful since the industry is often alleged to have committed itself at a panel conference.

The Advisory Council Committee report, previously mentioned, contained several recommendations to improve industry representation and permit a more constructive and representative panel conference:

(1) Widespread public notification at least 60 days in advance with efforts made to secure representation of large and small companies with geographic spread.

(2) More active participation of representatives of the Bureau of

the Budget.

(3) Preparation and distribution to participants of a formal

record of proceedings with opportunity for review and comments.

(4) Wider scope of subjects including whether or not a deter-

mination or redetermination is appropriate.

The Committee also recommended that these meetings not be "pro-forma" but be made truly advisory. Once survey data had been collected, it was suggested that the panel should be reconvened to further discuss the manner in which survey data is to be compiled.71 The following elements are likely to be decided upon either at the panel meeting (or at a later time on the basis of the presentations made at the panel meeting).

1. Tentative Definition of the Industry

At the panel meeting the Department of Labor proposes a definition of the industry. The definition is frequently based on the Standard Industrial Classification Manual which classifies establishments by type of activity.72 An industry is thus a grouping of establishments primarily engaged in the same line or similar lines

of activity. The line of activity is generally defined in terms of products made, materials consumed, or process of manufacture used.73

Major industries are classified by a 3-digit code, industry groups by a 3-digit code, and industries by a 4-digit code.74 Determination proceedings usually comprise one or more 4-digit industries; however, there are occasional deviations from this practice. Where actual practices of an industry deviate from the manual definition, certain items may be included or excluded.75 Items which are not purchased by the government are also occasionally excluded from the tentative definition1

Since the tentative industry definition developed at the panel meeting actually fixes the scope of the wage survey (and in consequence, the scope of the industry at the hearing), the importance of technically and economically competent industry spokesmen on the definition question at the panel meeting is evident.

2. Nature of the Wage Survey

In making a wage survey, the Department covers not only the establishments having government contracts but attempts to include all establishments in the industry which meet the size and sales cutoff limitation.

a. The Confidentiality Issue. Wage surveys are currently being made by the Bureau of Labor Statistics for the Wage and Hour and Public Contracts Divisions. However, the advisability of having BLS do the survey has been questioned because of the confidentiality issue, which will be discussed later.

b. Employment Cut-off.

In its wage surveys, the Bureau of Labor Statistics excludes from consideration small establishments

with a total number of employees below a certain "cut-off" point. Cut-off points of 5, 10, or even 20 employees are frequently discussed and adopted at the panel meeting.

The selection of the cut-off point is important because in most industries the pay practices of smaller establishments differ from those in larger establishments. For a truly representative wage survey, all establishments to which a determination is to be applied would have to be surveyed and a cut-off point higher than eight employees may well exclude from the survey some establishments which will have to pay the minimum wage determined.

In most industries, the establishments with the lowest minimum wages are usually small. On the whole, too, small establishments tend to pay lower than average minimum wages.

c. Branch or Product Determinations. Separate determinations are sometimes requested for branches of certain industries or specific product groups. Because such separate determinations have to be based on separate wage data, it is important that the need for such wage data be agreed upon at the panel meeting and before the survey actually takes place. If the opportunity to ask for separate wage data is missed, industry has to go into the hearings either without wage data to support its case for branch determinations or with costly wage survey and other supporting data of its own which it must seek to have introduced in evidence.

The Secretary is likely to make separate branch or product determinations when selected product groups differ from the rest of the industry by:17

(1) Different manufacturing processes; (2) Different Standard Industrial Classification Manual classifica-

tions;

(3) Different skill requirements; (4) Substantial differences in minimum wage levels; (5) Absence of competition; (6) Production in different establishments.

A demand for branch or product determinations, if successful: is likely to lead not only to one or more determinations which are

7 Tentative Labor Dep't Determinations: 25 Fed. Reg. 7801

(1960) (electron tubes and related products industry); 26 Fed. Reg. 5626, 7699, 8316 (1961) (paper and pulp industry); 25 Fed. Reg. 9903 (1960) (paper and paperboard containers and packaging products industry); 27 Fed. Reg. 1913 (1962) (motors and generators industry).

lower than a single industry determination would have been, but also to one or more determinations which are higher. Thus, in the recent Motors and Generators case, the midpoint of the medians In of lowest wages actually paid for the entire industry was $1.605. a branch breakdown, the midpoint for fractional horsepower motors was $1.48; the midpoint for nonfractional horsepower motors, however, was $1.725.78

d. Payroll Period. The payroll period for which the data are to be collected is also brought up for discussion at the panel meeting. The selection of the payroll period is of significance because the Secretary, as we have seen, is likely to ignore the lowest established rates in force unless they are also actually paid during the payroll period. Selection of a period which coincides with low employment is therefore likely to lead to the reporting of wage data which show as "minimum wages" the wages paid to employees who are above the lowest established rates.

To avoid this inflationary effect, industry should strive to have selected a typical payroll period during which there are substantial new hirings at the lowest established rates. To attain this objective, in case the current employment situation is poor, the survey date may be moved forward into the future or back into the past. If it is pushed backward, a concession may have to be made to take into account wage increases since the survey date; if it is pushed into the future, the survey itself will have to be postponed until such data are available. The Secretary has shown a marked reluctance in the past to move the survey date more than a few months either way irrespective of employment conditions in the industry. e. The 50 Per Cent Sales Test. The Department survey covers only establishments in which the manufacture of the products of the industry as defined constitutes 50 per cent or more of the total value of sales. Several industries have objected to this sales test as eliminating substantial manufacturers in the industry from being considered for survey purposes. Also, the 50 per cent sales test differs from the "primary product" test applied by the Census Bureau and other government agencies (including the Department of Labor) and thus makes comparison of BLS survey results with other economic data difficult. The Wage and Hour and Public Contracts

Divisions, however, hold that "the 50 per cent sales test is the best in this type proceeding and will, therefore, be applied."79

The 50 per cent sales test may, in some industries, exclude some of the largest manufacturers from coverage by the wage survey although such manufacturers are subject to the act. Where such an obvious injustice is likely, reconsideration of the Division's position should be requested. Unfortunately, it is usually impossible for most industries to determine in advance how harmful the 50 per cent sales test might be.

f. Lowest Established Rates. The need for data on lowest established rates has been mentioned and will be discussed in detail later.

Recent attempts by industry to secure inclusion of questions on lowest established rates in the survey questionnaire have been rejected by the Secretary at every panel meeting. Unless and until this attitude changes or is modified, it may be necessary for an industry to conduct its own separate survey covering, among other things, data on lowest established rates in force.

g. Covered Employees. Section 36 of Rulings and Interpretations No. 3, Walsh-Healey Public Contracts Act, states that the stipulations "shall be deemed applicable only to employees engaged in or connected with the manufacture, fabrication, assembling, handling, supervision, or shipment of materials ...and shall not be deemed applicable to employees performing only office or custodial work nor to any employee employed in a bona fide executive, administrative, or professional capacity ....

"So

Section 35 says that the determination as to whether a specific employee is subject to the act is a question for decision by the Department of Labor, while section 37 lists employees in particular occupations covered by the act.8' The list, unfortunately, is not complete or current. Moreover, there are no other readily available sources where all previous rulings on covered employees can be found.

The vagueness of the definition of covered employees, the limited guidance given in section 37 and the rather careless formulation in the BLS questionnaires of who is and who is not covered, have

led to very serious complications in past proceedings. It has, indeed, cast doubt on the probative value of all past wage data surveys. Up to a short time ago, the Divisions traditionally defined

"workers covered" in the questionnaire as follows:

For the purpose of this survey, include all working foremen and supervisory workers (including beginners) engaged in fabrication, processing, inspecting, receiving, storing, handling, packing, warehousing, and shipping; and janitors working around machines while in operation.82

This definition spoke of "workers" instead of "employees"; it said only "engaged in" and not "engaged in or connected with" as section 36 does; and it used terms different from section 36 ip describing the activities covered. Worse than that, this same definition excluded "other plant workers not mentioned above," thus excluding all those who had erroneously been omitted in the Department's deficient definition.

In two recent cases,ss the shortcomings of this definition were extensively discussed in the hearings. Witnesses produced evidence which demonstrated that the deficient definition had led to the exclusion of covered employees whose minimum wages were substantially below those elicited by the BLS survey. In the tentative decision in the machine tools case, the Secretary recognized the merit of industry criticism that blueprint machine operators and draftsmen and other "production-connected" jobs had erroneously been considered as outside the scope of the survey by persons answering the deficient questionnaire. In the tentative decision for the motors and generators industry, the Secretary admitted that some discrepancies between survey responses and industry affidavits and testimony "could have stemmed from misconceptions of the meaning of covered workers."'4 More importantly, recent BLS questionnaire forms contain a vastly improved definition of employees covered and employees excluded.85

(1957).

These recent developments confirm that industry representatives at the panel meeting should exert every effort to make certain that the lowest-rated covered employees, including those which are not plant workers, are clearly and specifically included by the definition printed in the BLS questionnaire. One industry recently prepared and secured the Department's acceptance of a specific list of job titles of "covered employees." This will aid those filling out the BLS questionnaires in deciding which employees were covered and which were not. In view of the potential importance of a minimum wage, such lists tailored to industry practices and peculiarities should be standard procedure.

h. Beginners or Probationary Employees. Industries with beginners or probationary employees can request "tolerances" or subminimums under which such employees may be employed for limited periods at lower rates. Beginners tolerances granted are generally 5 to 10 cents lower than the determinations, and are in force from one to three months' time. However, industries must present requests for special rates for such employees at the panel conference. If they do not, the wage survey may not separate beginner rates from those of nonbeginners, and it will be difficult, if not impossible, to obtain special consideration for beginners at the hearing. s

The fact that under current practice beginners are likely to be omitted from consideration if the survey period occurs during a period of stagnation or employment decline when no beginners are employed may have serious effects on an industry. In the case of the New England machine tool manufacturers, the Secretary's practice may threaten the continued existence of the industry in that region.

i. Apprentices. An apprentice, for the purpose of wage surveys, has been defined as a person who is employed under a formal written apprenticeship agreement which provides for a training period of not less than 4,000 hours.87 Such apprentices are excluded from the survey and from application of the minimum wage. This description has led to considerable confusion in some industries because it omits reference to or a summary of section 521.5, Code of Federal Regulations,"" which gives a much more detailed and In narrower definition than the one printed in the questionnaire. the machine tool industry survey, for instance, some apprentices employed under the industry's apprenticeship program apparently conformed to the questionnaire's definition as stated in the questionnaire. Since the respondents assumed their apprentices' wages were to be excluded, their lower wages were not reflected in the proposed determination. However, these same apprentices did not actually qualify under the Department's current interpretation of section 521.5, which is the standard used for determining whether or not an employee qualifies as an "apprentice." Hence, the higher minimum would apparently have applied to them.

The Department orally acknowledged the possibility of confusion and has expressed a willingness to clarify this definition. But the BLS has refused to permit any cross references to regulations in its questionnaires. As a consequence, the same confusing and incomplete definition remains in all questionnaires.

Industry representatives at the panel must therefore be fully familiar with the apprenticeship programs in the industry, and with the requirements of section 521.5. If the possibility of confusion exists, industry must try to inform those who fill out the questionnaires of the pitfalls contained in them.

j. Other Items in the Wage Survey. Other factors are occasion-

ally considered appropriate subjects of study in wage surveys. Each such question may raise special problems for industry.

In case there is a substantial time lag between the payroll period studied and the date of the survey, questions in regard to wage increases between the two dates may well be suggested by labor or the government. There have also been questions in regard to deferred contractual wage increases.

In recent proceedings, the Divisions have suggested questions as to the occupation of lowest paid employees. It has been agreed and stipulated in the latest questionnaire that the sole purpose of such a question is to permit verification that the lowest paid employee reported is truly "covered."

B. List of Establishments in the Industry

The basic source of names of establishments to be surveyed by BLS are the unemployment compensation records maintained by the different state agencies. These records are classified by industry. The structure of classification used89 is based on the Standard Industrial Classification (SIC) Manual (1957) which, as we have seen, is frequently also the basis for the proposed definition of the industry.

The Bureau of Labor Statistics obtains from the state agencies unemployment compensation tapes which list the names and addresses of the establishments for those SIC codes which BLS wants to survey. In addition, the Divisions usually ask the industry and labor representatives at the panel meeting to furnish lists of establishments which are available to them. Directories published by trade associations, commercial publishers, and regional groups, as well as classified telephone directories, are also consulted on occasion. Industry usually provides the Divisions or BLS with whatever lists are at its disposal.

C. Effectiveness of the Panel Conference

The panel conference, although considered "informal" is, in many ways, more important than the formal hearing. The shape of the wage survey is largely decided by the course of the panel conference. And it is the wage survey, more than any other one item,

of labor economics, particularly with respect to the collection and analysis of data on employment and manpower, productivity, and technological developments, wages, etc.93 The Bureau has 5 field offices, more than 1,000 employees, and a budget in excess of $10 million.

The Bureau of Labor Statistics, in undertaking a wage survey in Walsh-Healey proceedings, acts to some extent like an independent contractor in its relations with the Public Contracts Division. It follows the "client's" instructions in regard to definitions, questions to be asked, and tables to be prepared. It reserves for itself those prerogatives which assure the statistical integrity of the survey and the confidentiality on which the Bureau insists.

The general competence of the Bureau of Labor Statistics in the field of wage surveys has been largely accepted. There are several areas, however, concerning minimum wage surveys, which have led to criticism of the Bureau's role.

a. Confidentiality v. Right of Cross Examination. Wage surveys undertaken by the Bureau of Labor Statistics contain a confidentiality clause. This generally reads: "The replies will be treated in confidence. The completed questionnaire forms will be seen only by sworn employees of the Bureau of Labor Statistics and no information by company name will be released."95

The confidentiality clause effectively prevents industry, labor, and all Labor Department personnel (except BLS employees) from having access to the data underlying the wage survey. Actually, the confidentiality clause is interpreted so rigorously that BLS refuses to make available even a list of names and addresses of establishments which have been included in the wage survey. Some industries have held that production of all questionnaire responses is required by the Administrative Procedure Act" and applicable decisions in order to assure industry of its right to cross-examination.97 In a recent case,98 industry counsel requested that industry be furnished at least the names of establishments which participated in the survey. This request was rejected, primarily because of the position of the BLS,

even though such a revelation could have scarcely prejudiced the stated objective of the pledge.

The BLS plea of confidentiality is not based on any circumstances connected with Walsh-Healey proceedings. There are other avenues open to the Secretary for the gathering and compilation of data through which he could avoid the confidentiality issue which apparently is inherent in BLS surveys of this type.

Not all industries are in favor of making the individual establishment responses available for cross-examination. However, the decision in Powhatan Mining Co. v. Ickes," suggests that the question might be resolved in favor of forcing the BLS to disclose the data. In a recent case initiated by the Motors and Generators Industry9a and based on this issue the plaintiffs sought to enjoin the enforcing of the wage determination. The court citing Powhatan Mining Co. granted summary judgment for the plaintiffs, holding the determination null and void and enjoining its enforcement. At this writing the Department of Labor is in the process of appealing the district court's decision.

In the meantime, all efforts to get access to the names of respondents or to the responses themselves have been unsuccessful. The unavailability of the data underying the BLS wage survey is one of several reasons why industry often requests respondent establishments to furnish counsel with copies of their responses to BLS. In some instances, industries undertake special surveys of their own Industry can in addition to obtaining copies of BLS responses. then present its case on the basis of the data which it has been able In most to obtain directly from establishments in the industry. cases, however, industry is not able to obtain as many responses as BLS and the Department is disposed to give industry surveys less weight than BLS surveys.

b. Validity of BLS Industry Lists. The second area of concern is the doubt in the unemployment compensation tapes as the basic source of a valid list of establishments in an industry. The Bureau of Labor Statistics does not prepare the lists of establishments in an industry which is to be surveyed. To get the names of establishments within the scope of definition, it relies on unemployment compensation listing tapes from various states and lists furnished by industry and labor. In the Office, Computing, and Accounting

Machines industry, for instance, the names of 180 firms were obtained from the tapes. Eleven additional names were found in lists furnished by the International Association of Machinists and six additional ones in lists furnished by the Office Equipment Manufacturers Institute. During the course of the survey, six additional names were found. Thus, a total list of 203 establishments was contacted. A total of 97 were found to be within the scope of the industry and used for the wage survey.00

The responsibility for industrial classification for the unemployment insurance records is in the hands of state employment security agencies. While these agencies are instructed to follow the "Handbook on Industrial Classification," there is no assurance as to the accuracy of the classification process in the states. And since industry counsel does not have access to these records or to the list of respondents, this accuracy or lack of it cannot be measured.

In the authors' view, serious consideration should therefore be given to a change in the source of names of establishments. At least, a cross-check between establishment names available to the Census Bureau and unemployment compensation tapes should be instituted. Current weaknesses in the Census of Manufacturers series are not based on the unavailability of data. The weaknesses could be cured by greater emphasis on correct data for establishments in the smaller employment sizes, especially those with less than 20 employees. Only full publication of the names of all establishments included in a wage survey would assure industry's ability to check the validity of the survey.

2. Survey Practice

Surveys of wages have been made by the Bureau of Labor Statistics since 1888.101 A "Wage Studies Manual of Procedures" is used by Bureau employees which prescribes techniques of "universe" preparation, sampling, assignment of weights (in sampling), collection of data, editing, and coding.02

The Bureau of Labor Statistics depends on the voluntary co-

operation of the respondents, but is not above using pressure to ge cooperation. For this reason, there are only occasionally a feii establishments which refuse to fill out questionnaires. In the Office Computing, and Accounting Machines industry case103 one establish ment refused to furnish information, and one gave incomplete data which did not permit tabulation; this gave 95 usable replies and twc establishments that were "in scope" but could not be used.04 In a case of this kind, the BLS "weights" the data for the missing establishments of their location, size, and industry characteristics in accordance with standard statistical procedures and adds them to the available data.05

In most Walsh-Healey cases, the Bureau endeavors to make complete censuses of the establishments in an industry. On occasion, however, the Bureau relies on samples. The aim of sampling is to obtain sufficiently accurate surveys at minimum cost. The entire group of establishments which the survey results cover is defined as the "universe." The use of the sample is to provide estimates of the characteristics of the universe.10

In order to obtain maximum accuracy, the sample will be distributed among different size establishments in accordance with the principle of "optimum allocation." Under this principle, every plant has a chance of being included in the sample in approximate proportion to its size; for example, if there are 10 plants of about 200 workers and 10 plants of about 50 workers, the resulting sample, regardless of its size, will have four times as many plants of 200 as plants of 50 workers. Every one of the largest plants will be included. Since the sample will contain a greater proportion of large establishments than is found in the universe, a direct combination of the sample schedules would be biased in the direction of larger firms. To prevent this, numbers, known as "weights" are assigned to each sample member. Commonly, the weight assigned will depend on the chance of a given establishment being selected in the sample, Thus, a large establishment selected "with certainty" will receive a weight of one. If an establishment is selected as representing 5 establishments besides itself, it will be given a weight of 6.

Before the completed survey is tabulated, the Divisions prepare tables which show the proposed arrangement of the "dummy" tables but without statistical data.107 These "dummy" tables are sent to industry counsel and labor for comments. At that time, the industry has another opportunity to request the preparation of regional breakdowns, product breakdowns, etc., provided, of course, that questionnaire responses make such breakdowns possible. The

Divisions' willingness to grant industry (or labor) requests for the inclusion of questions in the questionnaire or preparation of special tables or breakdowns has varied over the years. Of late, they have been uncooperative.

Wage survey tabulations are usually made available to participants at the time of the notice of hearing or about 30 days in advance of the hearing. Thus, industry counsel will have only a few weeks in which to analyze complex tables, compare them with his own or other data, check possible errors with industry respondents, and prepare his case. Since counsel may be dealing with hundreds of respondents in many parts of the country, the 30 day period is rarely sufficient for the job which should be done.

E. The Hearing

Notice of hearing, pursuant to section 4 of the Administrative Procedure Act'08 is published in the-Federal Register, usually about 30 days before the hearing. The notice contains a tentative definition of the industry and informs "interested persons" that they may submit evidence restricted to a number of subjects and issues specifically stated in the notice. These generally refer to the definition, the question of branch or regional determinations, minimum wages, whether there should be a provision for beginners, and the effective date of a determination.09

The hearing, usually held at the Labor Department in Washington, is presided over by a hearing examiner appointed under section 11 of the Administrative Procedure Act.110 The examiner rules on admissibility of evidence but does not make determinations of fact or recommendations. He merely certifies the record as true and correct to the Secretary of Labor. All witnesses are subject to crossexamination, redirect, and recross-examination.

The first government witness presents data relating to competition for government contracts subject to the Public Contracts Act. These data show origin and destination of bids and awards for nonclassified government contracts. The purpose is usually to show that such competition is nationwide, thus in the Secretary's view,

requiring a single national determination. A witness from the Bureau of Labor Statistics next presents the wage survey and discusses the survey procedures and the tables. In an effort to prove wage increases between the payroll period of the wage survey and a time closer to the date of the hearing, the government usually produces a witness from the Divisions. The basic evidentiary document is a table showing increases in adjusted straight-time average hourly earnings for the period. The Divisions try to establish some relation between trends in adjusted straight-time average hourly earnings and minimum wages during that period.

As late as 1958, the Secretary rejected attempts of this type to connect changes in minimum wages with changes in earnings. He said:

No method is suggested to measure movements in prevailing minimum wages solely by movements in adjusted average hourly earnings, which are composed of such factors, unrelated to increase in minimum wages as variations in the amount of shift work, employment as between high wage and other plants, relative number of workers paid above the minimum, and others.1'

More recently, however, the Secretary has accepted the reasoning which he had previously rejected with such cogent arguments. In a recent tentative decision, he explains his current practice as follows: "I have concluded ...that the prevailing minimum wage may reasonably be assumed to have increased by at least the amount indicated by application of the percentage increase in average hourly earnings .... .112

Industry customarily presents its case after the government, and before Labor. It has frequently presented and sought to introduce in evidence wage surveys in support of its case which are more up-to-date than the BLS survey and include information on subjects not covered by that survey. Testimony on such wage surveys is generally presented by an economic consultant to the industry who usually has also been responsible for the conduct of the survey. This expert may also discuss the advantages and shortcomings of

different wage determination techniques. Other industry witnesses may present testimony and exhibits on manufacturing, wage, marketing, and other practices of the industry, depending on the specific circumstances.

While the labor presentation also varies, there are certain elements which appear often enough to permit the description of certain patterns. The first presentation is usually made by a Washington economist from the staff of AFL-CIO headquarters (if the interested unions are part of that labor organization). The economist analyzes the BLS wage survey and recommends a determination based on the median of lowest wages actually paid by establishments weighted only by covered worker employment (and not by the number of establishments). He also recommends that, in finding the median, the Secretary should treat establishments in which 1 per cent or less of all covered employees receive the minimum as if none received it. This means that minimum wages are counted only if received by more than 1 per cent of ihe covered employees. These two proposals permit the economist to suggest a much higher determination than would even be indicated under the current practice of the Secretary.

In regard to wage increases, the union spokesman usually suggests use of the adjusted average straight-time hourly earnings table introduced by the Department. Frequently, the economist or other union witnesses submit analyses of contractual wage increases granted after the wage survey.

At a recent hearing,13 the union also proposed the inclusion of fringe benefits in the determination. The Secretary in his tentative decision rejected this request, explaining that "such benefits cannot be viewed as part of a 'prevailing minimum wage' rate.114 In a subsequent hearing, the hearing examiner refused to admit union evidence of fringe benefits."5

The time granted for preparation of briefs, following the hearing, was formerly 30 days after receipt of the transcript. It has recently been shortened to 30 days regardless of the time when the tran.

script is available.116 At the same time, the Department stopped the preparation and free distribution of hearing transcripts,117 forcing participants to purchase transcripts from a contract reporter. The new practice imposes a heavy financial burden on industry and also causes a serious handicap in the preparation of briefs and of studies because the price limits the number of transcripts available.

After the hearing has been closed, no further evidence can be taken except at the request.of the Secretary or if provision has been made at the hearing for the later receipt of such evidence.

F. Tentative Decision

Under the Rules of Practice and the Administrative Procedure Act,"8 a tentative decision is usually issued and published in the Federal Register.119 Tentative decisions are usually fairly short, about four pages being typical. They rarely contain logical argument and often draw abrupt conclusions without attempting a justification or explanation. 1. Definitions (Including Product Divisions)12

Testimony at the hearing may lead the Secretary to reevaluate the tentative definition. However, changes in the definition are infrequent, especially if the formulation of the wage questionnaire does not permit any adjustment of the wage data to such a change. Administrative convenience seems to be the dominant rule.

Thus, the Secretary, in the tentative decision for the machine tool industry, rejected an industry request to include "exotic" machine tools because "the record does not contain sufficient information concerning the wages paid to employees engaged in the manufacture of these machine tools ... ."1 In like manner, the Secretary elim. inated television picture tubes from the tentative determination of the electron tubes industry because, among other reasons, "the data in evidence permit determination for the remainder of this industry with these products excluded."'22 In permitting two product de-

110 Secretary Announces Improvements in Wage Determination Procedures, Dep't

of Labor News Release No. 4774, Sept. 18, 1961; 41 C.F.R. § 50-203.21 (a) (1962 rev.).

127 41 C.F.R. § 50-203.17 (c) (1962 rev.). 11860 Stat. 242 (1946), 5 U.S.C. § 1007 (b) 11 41 C.F.R. § 50-203.21 (b) (1962 rev.). 120 The factors entering into the formulation of the definition have already been

discussed, supra at 223-24.

12127 Fed. Reg. 898, 899 (1962). 1 125 Fed. Reg. 7801 (1960).

terminations in the motors and generators industry, the Secretary stated that because "there appears to be an appropriate basis... and additional time will not be required for such division, I have decided to make the separate branch determinations.... "128

The Secretary's practice emphasizes the importance of separate wage data in case of the need for a change in, or product breakdown of, the proposed definition. This in turn emphasizes the need for full preparedness of the industry at the time of the panel meeting so that requests for separate wage data may be made before the BLS questionnaire form is agreed upon.

2. Locality

The Secretary has taken the position that determinations are to be made on a nationwide or industrywide basis where an analysis of government procurement shows that bids are submitted, and contracts awarded, for delivery on a nationwide basis. In recent tentative decisions the Secretary has kept to this position even though the analysis prepared by the Divisions in one case was called a "history of errors, repeated errors, and general confusion about inclusion or exclusion of contracts representing as much as 50% ox more of the total sampling,"124 and in another was attacked because different criteria have been applied for inclusion of establishments in the origin and destination tables and in the wage tables.'25

3. Prevailing Minimum Wages

a. Continued Use of the Median Technique. In recent determinations, the Secretary has continued his application of a technique which finds the "prevailing minimum" wage to be one which is paid by approximately half or more of the plants surveyed, which employ more than half of all surveyed employees covered by the act. This is generally called the median technique.26 The median, in a group of lowest wages actually paid (and ranked according to

the amount actually paid), is that lowest wage that neither exceeds, nor is exceeded by, more than half of the observations.127 In brief, it is the middle value in any group of observations. Because the Secretary attaches significance to the median of the plants weighted by their covered worker employment and that of plants without such weighting, he has to find a prevailing minimum somewhere between the two medians.

A typical tentative decision arrives at a proposed minimum wage

as follows:

[T]he median level of the minimum wages paid by the plants involved was $1.45 .... The comparable level for the plants weighted by their employment was $1.60 at the same time. A simple average of the two medians results in a minimum wage of $1.525.128

However, the Secretary usually does not propose to set the "prevailing minimum" automatically at the mid-point between the two medians. He does look for a minimum between the two medians at which there are some plants with a substantial number of employees. Here is a quote from the same tentative decision:

However, an examination of the table reveals that the minimum wage of $1.525 has no other significance because there were no plants paying minimum wages above $1.51 and under $1.54 .... [T]he minimum wage of $1.50 appears to have particular significance. There is a substantial [T]he plants at the $1.50 point have cluster of plants at the wage .... more covered worker employment than appears at any other point in the critical range between the medians .... Therefore, I find $1.50 to be the prevailing minimum wage .... 12

b. Neglect of Lowest Established Rates. As early as 1950, the Secretary began to give some consideration to contract minimum rates even in industries where there was no truly prevailing minimum. Thus, in the determination for the aircraft manufacturing industry he stated: "The importance of the interval containing $1.05 is also indicated by the fact that a considerable number of union agreements include current minimum job classifications in this interval."'30 As late as 1957, the Secretary stated in his proposed

determination for the Scientific, Industrial, and Laboratory Instruments Industry that "consideration has also been given to the lowest established rates in plants in this industry .... ."81

The hesitant recognition by the Secretary of the significance of lowest established rates apparently came only after pressure from the Advisory Council on Federal Reports during 1949 and 1950. The Report of the Walsh-Healey Panel to the Budget Bureau of February 2, 1950 stated: "A minimum wage in an individual plant is the rate which has been established for the least skilled job in the plant."182 Shortly after 1957, the Secretary discontinued giving even lip service to the importance of lowest established rates. In the tentative decision for the Manifold Business Forms Industry, he stated flatly that "minimum wage rates that were established as a matter of policy but were not actually paid on the survey date ... should be .,8 This position, which has been reaffirmed in disregarded .. later tentative decisions, constitutes one of the major inequities in the present administration of the act. •

4. Beginners or Probationary Workers

Whenever the Secretary finds "widespread" or "fairly extensive" use of beginner or probationary worker pay practice, he provides a separate subminimum rate for such employees under section 6 of the act.184 In establishing the differential for beginners and probationary workers and the time period during which the subminimum may be paid, the Secretary frequently applies median techniques to tables showing lowest wages paid such employees, wage differentials, and time intervals.

It is the practically unlimited discretion permitted the Secretary under this section which the Secretary used to set a subminimum for beginners in the electric lamp industry at $1.20 when the facts in In this case, the the case indicated a determination at $1.00.185

Secretary ignored the table on lowest wages actually paid beginners and relied on a table showing established hiring rates, saying that there was "no proof that these rates had been or would be inoperative or that they were other than bona fide."'136 This, of course, is precisely the industry's argument for lowest established rates which the Secretary flatly rejects when it does not suit him.

5. Proposed Determination

The tentative decision concludes with a notice of proposed determination which includes the definition of the industry and minimum wages and gives "interested" persons permission to file written exceptions within 21 days from the date of publication of the tentative decision in the Federal Register.

G. Final Decision

Final decisions in the determination normally discuss the exceptions. Most frequently industry's arguments are summarily rejected:

Each of the exceptions has been carefully considered. None appears to warrant further discussion here. Each is overruled. Each finding and conclusion, together with the reason and basis therefor, which is expressed in the tentative decision, is hereby made final.137

Similar arbitrary rejection of industry exceptions by the Solicitor in a Davis-Bacon Act case elicited the following comment from the Comptroller General of the United States: "[W]e believe that (the regulations) also provide that the record show the basis for ruling on each significant exception so that the Department's position will be clarified for the record."'38

The effective date, which used to be 30 days after publication, has been cut to 7 days after publication in the Federal Register, "unless evidence introduced at the hearing shows that such action would be inappropriate in that case."'139 Industry has strongly objected to the seven-day period. It holds that the burden of justifying a departure from. the thirty-day period rests on the Secretary'4" and that

the imposition of the seven-day period will work a serious hardship because of the need to first prepare, print, and distribute copies, adjust accounting practices, determine a new company wage policy, determine whether to compete for government contracts, and finally to consider whether to present to the courts a plea for a stay of execution of the determination.141

IV

The Secretary's wage determinations have largely escaped the close and continued judicial scrutiny that would seem appropriate in matters of substantial import to the national economy. This can be ascribed primarily to four factors: (1) In the first decade of the act, most determinations did apply to industries in which sweatshop conditions existed; (2) delays in making and applying determinations based on earlier wage survey data coupled with rapidly increasing wages substantially reduced the economic impact on the industries (3) not until 1952 was judicial review of and regions in question; wage determinations available to aggrieved parties; (4) most industries have been poorly prepared to engage in the long drawn-out and costly processes required for judicial review.

In general, it can be said that the courts have sought to avoid entanglement in the economic and statistical complexities of minimum wage determinations and have sustained the Secretary in those few matters which have been brought before them. This has apparently had a most unfortunate effect on the Secretary in that the determination process, in the judgment of the authors has increasingly become a stereotyped affair wherein the Divisions, relying, almost solely on confidential government-collected and tabulated data, unilaterally establish single, nationwide minimums for American industries. This same disposition towards unilateral action has also marked the regulatory process pursued by the Secretary under the act. The promulgation of regulations is almost entirely without benefit of hearing and frequently without notice or opportunity to submit objections or suggested changes in any form. An especially

interesting example of regulatory horrors which might have benefited from the advice and participation of industry was the extensive and far-reaching safety regulations which became "law" in early 1961.142 The absurdities, both in phraseology and practical consequence, of this collection of rules-ranging from the dimensions of stair treads. to broad, sweeping generalizations-were catalogued in an elaborate monograph prepared and published by the Electronic Industries Association.43 The only apparent consequence of this analysis, which the Secretary has not attempted to refute, is that thoroughgoing enforcement has apparently been delayed. The real danger is that if these and other rulings remain unchallenged, they may be upheld on grounds of "consistent practice" and administrative convenience.

A. The Lukens' Period: No Standing to Sue

The first challenge to enforcement of the act came two years after its passage. A group of manufacturers in the iron and steel industry challenged a wage determination which divided the entire United States into six "localities" and established a minimum wage for each of these regional groupings. The manufacturers argued that it was a gross distortion of the statutory term "locality"'145 to construe it to mean a multi-state area encompassing as many as fourteen states and the District of Columbia.

The Court of Appeals for the District of Columbia sustained the 142 Safety and Health Standards for Federal Supply Contracts, 41 C.F.R. § 50-204

248 manufacturers' contention.14 It pointed out that although the word "locality" is a term of somewhat indefinite meaning, at times crossing the lines of fixed political subdivisions, "the determination in this case goes so far beyond any possible proper application of the word as to defeat its meaning and to constitute an attempt arbitrarily to disregard the statutory mandate."'147 The Supreme Court, however, without touching upon the proper meaning of the "locality" standard, reversed the court of appeals on the ground that no legal right of the manufacturers had been invaded and that they therefore had no standing to challenge the Secretary's action.48

B. The Fulbright Amendment: More Apparent than Real

Mounting congressional concern over the Secretary's power was reflected in the enactment in 1952 of the Fulbright Amendment, which reversed the effect of the Supreme Court decision in Lukens by providing for judicial review of wage determinations.14

1. The Death of Regional Determinations

Secretary of Labor Tobin, in the first determination after the amendment became effective, promulgated a national minimum wage of $1.00 for the textile industry.150 The district court held this determination invalid.15 The Secretary urged that the term "locality" was irrelevant, on the theory that the phrase in which it appeared-"currently operating in the locality"-related only to the immediately preceding alternative, namely, persons employed in groups of industries. Since he had rested his decision on the standard of persons employed on "similar work," rather than persons employed in "groups of industries," he contended that the "locality" mandate did not come into play. District Judge Holtzoff, noting that this interpretation was untenable and "more or less of an afterthought," refused to "place a tortured interpretation and attach a distorted meaning to a simple English word."152

In a split decision, the court of appeals discarded its earlier reasoning in the Lukens case as "without value as a precedent" and reversed the district court.153 Essentially holding that the disputed phrases of the act had no plain meaning, the majority of the court refused to upset this particular exercise of the Secretary's discretion. The dissenting opinion noted that nothing had occurred in the interim to disturb the court's opinion in Lukens, and cited legislative history to support the interpretation advanced by the aggrieved manufacturers.

2. Open Market Exception

Another facet of wage determinations was examined in Ruth Elkhorn Coals, Inc. v. Mitchell.154 There certain mining concerns sued to invalidate the wage determination applicable to the bituminous coal industry. Protesting that the application of the act to the coal industry contravened the express statutory exclusion of "purchases of such materials, supplies, articles, or equipment as may usually be bought in the open market,"'155 they reasoned that because bituminous coal is mined for general use and may be bought by the government in normal competitive channels, it is exempted by the open-market provision. The Secretary argued that the exemption applies only to purchases which the government itself is authorized to make in the open market. The narrow construction urged by the Secretary was adopted by the court.56

3. Regional Determinations Revisited

The rigidity of the judicial process in wage determination matters was made clear in a recent court test, Consolidated Electric Lamp Co. v. Mitchell.157 In this case, involving the electric lamp in-

dustry, the Secretary had shrugged off the industry's contention, among others, that the act limits the geographic scope of a wage determination to a locality, at least when it is based on wages in a particular industry. Although this case differed crucially from the textile determination in that here the Secretary relied upon wages in the electric lamp industry and in the textile decision the Secretary had relied upon wages paid for similar work, the court of appeals perceived no distinction. In a one paragraph opinion citing Covington Mills as controlling authority, the court affirmed the Secretary.

The apparent unanimity manifested in the Lamp case is at odds with the earlier independent opinions of three district court judges'58 and three circuit court judges19 who held that the word "locality" cannot include the entire United States.

C. Possible Alternatives

Part of the weakness in judicial review may lie in the fact that it is confined to the courts of the District of Columbia, the jurisdiction in which the Secretary resides. The court's deference to the Secretary's interpretations and determinations has effectively discouraged aggrieved parties from seeking judicial review. The appellate procedures opened by the Fulbright Amendment have thus been more apparent than real.

One solution might be to adopt a procedure similar to the one under the National Labor Relations Act, where an aggrieved party may sue in the jurisdiction in which his aggrievement occurs, i.e., the judicial circuit in which his manufacturing establishment is situated.00 In this fashion other courts of appeals would be authorized to examine the legality of nationwide minima, as well as other doubtful standards advanced by the Secretary in wage determination proceedings. Such decentralization of the judicial process would give rise to a continuing scrutiny of administrative decisions in this field, and through the likely diversity of opinion, might spur Supreme Court resolution of the basic issues.

A further reform of considerable merit was advanced in 1958 by the House of Delegates of the American Bar Association,161 which adopted a resolution to the effect that judicial review be initiated in the United States Courts of Appeals rather than the district courts. This procedure, following the customary method of review of administrative decisions, would hasten final decisions concerning the legality of the Secretary's determinations, and the sufficiency of the evidence on which they are based.

The development of appellate permissiveness towards the Secretary's actions has been to intensify appeals to the Secretary himself. As the Secretary issues the initial determinations to which briefs of exceptions customarily are filed, the Secretary is put in the position of being asked to confess error in modifying his determinations. Perhaps a more satisfactory procedure would be for the initial decision to be issued by the hearing examiner who conducted the hearing. The independent and impartial aspects of the hearing examiner's role should be continually fostered so that the initial determination in the case would be the result of intimate familiarity with the facts, made by a man insulated by independent status from an attitude of "team spirit." This arrangement would permit a genuinely fresh and objective review by the Secretary's personal staff, and preserve to the Secretary freedom to express administrative policy on the basis of data independently and objectively assembled and interpreted by the hearing examiner.

V

EFFECT OF DETERMINATIONS A. Available Data Limited

The effect of prevailing minimum wage determinations under the Walsh-Healey Act has been largely ignored by social scientists. Statistical data on the effect of such determinations are practically nonexistent. The 1962 Budget of the United States Government promised that "during 1962 a pilot study of the economic effects of Public Contracts Act wage determinations will be conducted.162 Meetings of representatives of the Divisions, the Bureau of the Budget and the Advisory Council on Federal Reports, were held late in 1962 and early in 1963 to formulate such an impact study.

The Walsh-Healey Committee of the Chamber of Commerce has recently presented eleven recommendations to the Bureau of Budget on ways to insure a meaningful study.

More extensive data are on hand on the effects of minimum wage determinations under the Fair Labor Standards Act,13 and further material was submitted in January 1963. On the whole, these studies tend to show that, immediately after the new FLSA minimum became effective in low-wage industries, average hourly earnings increased substantially, a larger proportion of workers earned just the minimum wage, the dispersion of workers' earnings narrowed as the new minimum raised the floor under wages, and the wage differIn entials between higher and lower paid workers were reduced. the longer run, some restoration of the industry's wage structure as it existed prior to the increase in the minimum was noted, but the degree of restoration varied widely by industry.164 We do not know if wage increases due to Walsh-Healey determinations have a similar effect.

B. Immediate Direct Impact of Walsh-Healey Determinations The statistical tables prepared by the Bureau of Labor Statistics in Walsh-Healey cases show how many employees get wages which are below the level ultimately determined as the prevailing one by the Secretary. Because these tables show how many employees would be affected by a determination if their employer were to bid for government business (and if their wages had not changed in the period between the wage survey and the effective date of the determination), these tables have been called "impact tables."

A summary presented before the Senate Committee on Banking and Currency in May 1952 shows that between 2.3 and 9.0 per cent of all covered employees in 14 industries with postwar determinations were earning less at the time of the wage survey for the industry than the "prevailing minimum."15

In more current cases, the "impact" is close to, or larger than, In the machine tool industry, for that of earlier determinations. (before a 7 instance, the Secretary proposed a minimum of $1.73 cents allowance for wage increases). Four per cent of all covered

employees were paid less than the proposed minimum at the time of the wage survey for that industry. In the fractional horsepower motor and generator industry, the comparable percentage was 11.5 (at $1.45); in nonfractional, 6.3 (at $1.70).166

The effective implementation of the "speed-up" regulations would, of course, increase the impact of any more recent determination.

C. Adjustments of Wages in Higher Brackets

Industry spokesmen have emphasized for some time that wage adjustments to a new Walsh-Healey minimum tend to spread to other parts of the wage structure. The cause for this unavoidable increase in wages not effected directly by a determination was believed to be the need to maintain established differentials between different classes of jobs involving different degrees of skill, effort, and training.67

A recent survey by the National Machine Tool Builders' Association of 123 plants gives statistical confirmation to this analysis. While the expected direct first-year cost to adjust to an $1.80 minimum was found to be $1.8 million, the total first-year cost "to maintain differentials where believed essential" was $8.5 million.68 The cost of adjustment was thus several times higher than the direct immediate cost.

D. Regional Impact

It is obvious that the "impact" on certain regions and certain size groups of plants will be much more severe than is indicated by industrywide averages. An analysis of wage determinations made between 1948 and 1950 showed that there were from 1 to 11 per cent of all employees below the minimum set by the Secretary on a nationwide basis; when the region most seriously affected in the identical industries was analyzed, the percentages varied from 4.8 to 71.0 per centl9

The Machine Tool Builders' study, previously referred to, found that the proposed determination at $1.80 would hit the New England states hardest. It was found that in New England, 98 per cent

of all employees subject to the new determination work in responding plants which paid less than the proposed minimum. To be in a position to qualify for government business, the respondents reported.that they would have to add more than $340 in increased payroll cost for each covered employee per year. Also reporting as being hard hit if such a minimum were implemented were the South and New York, New Jersey, and Pennsylvania.170

It is apparent that the wage determination practice of the Secretary works to the serious detriment of the older industrialized areas of the country and of those just starting the process of industrialization. Representatives of New England who initially endorsed the Walsh-Healey Act in the mistaken belief that it would help their region71' have seen in the Machine Tool case that this is not so. It is largely well-established and well-equipped plants in the North Central and Western states which will not be immediately affected adversely by the current determination practice. The continued existence of older plants and the growth possibility of undeveloped regions is thus put into jeopardy by their being handicapped in competing for government business.

E. Effect on Individual Plants

The large majority of plants most seriously affected by a determination (or excluded from government contracts) are the small ones.'72 In the machine tool industry the 20 lowest paying plants (compared with an industry average of averaged 77 employees 148);17a in the electric lamp industry, the 10 lowest paying plants (compared with an industry average of averaged 86 employees 257);174 in the motors and generators industry the 20 lowest paying plants averaged 116 employees (compared with an industry average

of 240).175 In the drugs and medicine industry, where the concentration of production in large plants was very evident, the Secretary made a determination which affected less than 7 per cent of the largest plants but more than 70 per cent of the smaller establishments.176

The handicap which minimum wage determinations place on smaller establishments may not be due to their paying lower miniIt may well be due to the mum wages for the same kind of work. fact that the characteristics of the lowest job differ between plants in the same industry. If this is correct, the mechanistic-statistical approach currently used by the Secretary may penalize many establishments by disregarding such differences.

An analysis of the plants on the basis of capital equipment, hiring practices, etc. in any given industry generally indicates a pattern. While patterns naturally vary from industry to industry-and even within components of industries-the general picture is likely to be as follows.

At the bottom, from point of view of minimum wages, will be small, poorly capitalized and poorly equipped plants. Wage rates in such plants tend to be low because of low skill requirements, low hiring standards, shortage of capital, low productivity and frequent It is in this bottom group that one absence of union organization. is likely to find the lowest minimum wages.

At a somewhat higher level are plants of all sizes with a sounder capital structure which operate with better, but not the most up-todate equipment. While skill requirements and productivity in such plants may be somewhat higher, they are still fairly low. Hiring standards are more selective; plants are more frequently unionized but minimum wage rates still tend to be somewhat lower than, or close to, the average for the type of job.

Overlapping this group to some extent in regard to rate structure are the plants of large companies with high capital investment per In worker. These plants are usually of large or medium size. many industries, the plants of the four largest companies of this type make up more than half of total shipments of the industry-

75 per cent of all motor vehicles, 54 per cent of all steel, 62 per cent of aircraft engines, 79 per cent of tires and tubes.177 These highly modernized plants will break their jobs down into more classifications. While hiring standards and productivity will be higher (depending, to some extent, on local characteristics of the labor force), the extensive job breakdowns will permit the establishment of comparatively low grades. Plants of this type are either unionized or pay rates equal to, or better, than unionized plants. Minimum wages will tend to be close to the average for the job-but on the higher side.

At the top of the wage scale are likely to be some small, highly specialized and well equipped craft-shops. The lowest rated workers in such jobs often perform both higher and low skill operations. Minimum wages, therefore, tend to be very high. They are obviously exceptional and as untypical of minimum wages for the industry as are the very lowest minima.

If the general picture given here is correct, we find that a few large companies and their plants dominate the minimum wage pattern. Any determination which places major emphasis on the minimum wages in the top half of the industry is likely to impose the minimum wages applicable to job classifications in modem plants with high investment per worker on plants in which the lowest-rated jobs are of a different nature."8 The effect will be compounded if the lowest wages paid in small craft-shops with exceptionally high earnings are also included in the Secretary's calculations while those with exceptionally low earnings are excluded from consideration.

F. Effect on Future Hiring

An extreme example of the effect of a determination based on the median technique applied to wages of highly paid senior employees was brought out in the exceptions filed by the machine tool industry and in the meeting of presidents of New England machine tool companies with the Secretary of Labor. The machine tool industry, just prior to the BLS survey period, was in such poor health that the situation was felt in some quarters to constitute a threat to national security. The industry was generally acknowl-

In 1960, edged to be operating at 50 to 60 per cent of capacity. only 45 out of 287 plants in the BLS survey had beginners; there were only 316 beginners reported during that period out of roughly 42,000 covered employees.179

In the many years of contraction, the average age of employees in this industry has increased to more than 50 years; an industry spokesman at the meeting with the Secretary pointed out that the industry has finally contracted. so far that it was beginning to hire a few new and junior employees again. Yet, if the proposed minimum were made effective, the industry would be faced with a determination without a beginner's tolerance, since at the time of the survey there were not enough beginners to meet the Secretary's requirements established for such tolerances. Established beginners rates in the plant of one of the industry's spokesmen, while amongst the highest paid in his state, were as much as 40 cents below the determination proposed by the Secretary.

G. Other Effects80

Minimum wage determinations under the Walsh-Healey Act apply only to work on government contracts. However, in many situations it is not possible to identify parts and products according to the ultimate consumer. And even if such a separation were possible, union contracts and job classification plans make it difficult to establish different pay scales for identical work merely because one product is shipped to the government while the other one goes to a private customer. The minimum wage rates set by the WalshHealey determination for an industry will therefore tend to apply with equal force to all work performed in a plant which has, or had, or wants to have, a government contract.

Only in those plants which do not have (and are not intending to bid for) government contracts is there a possibility of continuing operations at less than the wages found to be prevailing for the industry. But even in such plants there are many factors tending to eliminate .wage differentials between them and other plants in the same industry. Among these factors are nationwide collective bargaining, coverage of different plants by the same union, effects

of wage increases in one plant of the industry on other plants even if there is no union, and so on.

One of the most serious effects of a determination under the prevailing practice is that it starts a spiral of ever-increasing wages. Once the rates of all covered employees have been increased to the "prevailing minimum" set and the adjustments in wage differentials have been made for rates above the minimum, a redetermination on the basis of the new minimums actually paid would undoubtedly be higher. This is so because the wage rates previously below the minimum would have been raised to at least the minimum and other wage rates are likely to have been increased to maintain In addition, there can be little doubt that in some differentials. plants the lowest rates would temporarily be unused and the lowest minimum wage paid would be above the rate. From then on, the procedure could be repeated at regular intervals: wage survey, then determination, then immediate wage increases, then readjustments of the rate structure, then wage survey, then redetermination, then immediate wage increases, etc., etc.

The spiral of ever-increasing wages by administrative decision made possible by the present median technique is limited only by the capacity of the staffs of -the BLS and the Divisions to process such cases and by the effect which plants not engaged on government contracts have on the wage structure.

VI

A. Review Within the Framework of the Act

There can be little doubt that industry wants to see the WalshHealey Act repealed or drastically revised. Regardless of the personal view of the authors, this article is not concerned directly with legislative amendment to or revocation of the existing act. Rather, we have confined our analysis to needed changes in the administration of the act within the framework of the law, as amended.

B. Procedural and Organizational Shortcomings

1. The General Problem

Hearing records, briefs, and exceptions are replete with industry complaints of improper statistical procedure, lack of reliable evi-

dence, denial of rights, use of unsuitable methods, assumption of adversary role by department counsel, etc. Most of these complaints have been brushed aside, frequently in decisions which obviously have been written largely by the same counsel who considers the proceedings as adversary.181 The only concessions on a higher level that have been made in response to industry complaints have been a promise that "comments on tentative decisions will be reviewed in the Office of the Secretary by.persons other than those who participated in the hearings or in the drafting of the tentative decision"' and an announcement that the Secretary "will appoint a labormanagement advisory committee to help make certain the WalshHealey Act will be administered 'properly and equitably.' "182 To this day, no advisory committee has been appointed.

Most industry advisers feel that a review procedure which starts only after a tentative decision has been made in the name of the Secretary makes improvements in such a decision unlikely.83 This is evidenced by the final decision in the Motors and Generators case which rejects industry objections just as summarily as they were rejected in previous reviews-and again without stating any basis for such rejection.84 This is exactly the practice severely criticized by the Comptroller General in regard to similar practices of the Secretary in a Davis-Bacon case.:85

On the statistical side, the Bureau of the Budget and its Advisory Council have had little effect on the correction of statistical practices since their partly successful intervention in Walsh-Healey matters It can be hoped that with the recent submission of the in 1950.186 report of the Walsh-Healey Committee of the Advisory Council that important influence will again be constructively felt.

2. Proposed Remedies

This is not the proper forum to propose in detail ways for correcting the many additional major and minor shortcomings in the administration and statistical procedures now current. Rather, we shall outline briefly what existing and new instrumentalities and practices might best be considered for carrying out the urgently needed improvements.

a. The Panel. The authors agree with the recent recommendation of the Advisory Council's Committee on Walsh-Healey'87 that objective criteria should be established for selecting industries which are to be made the subject of a determination proceeding. This criteria should give full recognition to such factors as the level of employment, the relative economic health of the industry, the current or expected impact of foreign competition, the extent to which depressed or labor surplus areas will be affected, etc.

Interested parties (including such government agencies as the Departments of Defense and Commerce, the Bureau of the Budget, the Small Business Administration, etc.) should be permitted and invited to present evidence on the advisability of determination or redetermination proceedings for the industry within the framework of the objective criteria.

The industry definition and the content of the wage survey should also be the subject of the panel meeting. Formal requests for changes in, additions to, or deletions from, the definition or the questionnaire by interested parties should be given full consideration. Refusal to grant such requests should be justified in full in writing. Proceedings at panel meetings should be formal and complete records should be furnished to participants free of charge. Any decisions made on the basis of the panel meeting in regard to selection of an industry, definition, or wage survey should be subject to exceptions by participants.

b. Decision Making. Proposed decisions, including findings of fact and conclusions of law, should in the future be prepared by a qualified hearing officer or, possibly, by an independent panel which would preside over the hearing. Such proposed decisions should be signed only by the persons responsible for the drafting and not by the Secretary.

A Public Review Board, including at least one person familiar with industry practice in wage and hour administration, should review all proposed decisions. The Board should have power to approve such decisions or revise or reject them for cause. Final decisions by the Secretary should be issued only after the Public Review Board so recommends.

c. Review of Proceedings. There is an urgent need for greatly increased review and reexamination of Walsh-Healey practices. The promised appointment by the Secretary of Labor of a labor-management advisory committee to "help make certain the Walsh-Healey Act will be administered 'properly and equitably' "188 conforms to a suggestion expressed by industry more than 12 years ago.189 It is regrettable that the promise has not been carried out. The value of such a committee will depend on the tasks it will set for itself, on its composition, its ability to work smoothly, and on the degree to which the Secretary follows its advice.

to

Considerable authority

improve statistical procedures of Federal agencies is already vested in the Bureau of the Budget. The Office of Statistical Standards of that agency may "plan and promote the improvement, development, and coordination of Federal and other statistical services."''10 During 1949 and 1950 the Bureau, under the stimulus of the Advisory Council of Federal Reports, helped considerably in eliminating some of the most objectionable features of the then current practices in the Walsh-Healey administration. There is no reason why it could not again assume such a role.

Increased review of Walsh-Healey determinations by the Courts

would also contribute greatly to improved practices.

C. Statistical Shortcomings

1. The General Problem

The Secretary has reduced the minimum wage determination process to a simple statistical exercise. He analyzes the lowest wages actually paid and picks among them a wage which is being equalled or bettered by half the establishments in the industry with half or more of all employees subject to the act.

The major shortcomings in the Secretary's statistical approach are: the use of the median technique;191 the indiscriminate dependence on lowest wages actually paid;192 and the questionable handling 2911 The median technique bases the determination on the minimum wage practices It arbitrarily declares the practices of the top half to

of incentive earnings.193

In doing this, the Secretary ignores many economic facts of life and some statistical ones as well. He seriously disturbs the normal development of regional economies, the normal recovery of depressed industries, the ability of plants with less-than-average facilities to compete on an equitable basis, and other elements of economic progress. He does all this unnecessarily because none of these actions further the accomplishment of the purposes of the act; they only make the determination process easier for the Divisions' staff.

2. Proposed Remedies

In making a determination the Secretary should follow the

procedures set out below.

a. Replace the Median Technique by One Which Gives Proper Consideration to Normal Industry Practice. What we propose here is based on the simple consideration that "prevailing" practice can be found only by a study of the practice at and about the average, not by a study of only those plants which pay more than the average. While the statistical analysis in the footnote may be hard to follow, it really describes in statistical language a basic and simple technique.194

This approach, however, is not the only rational technique available to replace the faulty median technique. Another way to find a prevailing minimum might well be the application of the "first significant duster" technique to minimum wages. The major fault of the duster technique as used in the past was not any inherent fault in the technique itself but its patently erroneous application to tables showing the earnings of all covered employees in an industry. The cluster technique, properly limited to frequency distributions of

lowest wages paid or in force in an industry, would produce valid results.

b. Use Lowest Established OR Lowest Paid Wages, Depending on the Circumstances in the Case. It is apparent that neither the data on lowest wages actually paid alone nor the data on lowest established rates alone (if collected) are adequate for an equitable determination.

Several factors require elucidation to enable the Secretary to

make an equitable determination:

(a) Information respective of whether or not lowest established rates are bona fide rates which will be used whenever employment conditions warrant;

(b) Information on employment fluctuations in the plants of respondents to the questionnaire at specified payroll periods preceding the payroll period of the survey;

(c) Information on lowest established rates and on lowest wages

actually paid;

last (d) Information on when lowest established rates were used and, if not used at the time of the payroll period, for what specified reasons they were not used (e.g., no new hiring since time of last use, never used, hiring at higher rates).

Lowest wages actually paid may-or may not-be equitable as a basis for a determination in an industry which generally has at least stable employment and normal new hires. Lowest wages actually paid are not an equitable basis for a determination for any industry with already low or declining employment and abnormally low new hires. Only judicious use of full information on lowest established rates can lead to equitable determinations for such industries.

c. Make Proper Adjustments to Incentive Earnings. For incentive workers, it is proposed that the minimum hourly rate proper for use in determining the minimum is the guaranteed hourly minimum rate or a rate computed dividing hours worked into straight time earnings (excluding shift premiums) less 20 per cent, whichever is higher. The percentage may be adjusted to industry experience in the Walsh-Healey panel conference for each industry.

D. Conclusion

In this article, we have attempted to familiarize those who might become involved in Walsh-Healey determination proceedings with

the act and its administration. We have shown that industry is seriously concerned with the current administration of the act-and why this is so. Beyond this, we have tried to point to certain avenues for correcting the existing situation. Any true improvement in the administration of the act depends on the desire of the Secretary to make such an improvement or, as the last resort, on legislative relief.

PROBLEM CHILD AMONG LABOR LAWSTHE WALSH-HEALEY ACT

I. Introduction

A. The Reasons for Industry Concern

1. Change in Approach 2. Acceleration of Determination Process 3. Effect of Low-employment Periods

II. Background

A. History of the Act

1. The Roosevelt Era 2. Developments Since 1948

B. Interpretation and Implementation of the Act

1. Stipulations Required 2. Terminology of Section 1 (b) 3. What Is "Prevailing"

a. The Contract Minimum b. The Cluster Technique c. The Median Technique 4. What Are "Minimum Wages" 5. Whose "Minimum Wages" Are Studied

a. The Language of the Law b. The "Particular Industry" Standard

6. Locality 7. Open Market

C. Administration of the Act

III. The Determination Process

A. Labor-Management Panel Conference

1. Tentative Definition of the Industry 2. Nature of the Wage Survey a. The Confidentiality Issue b. Employment Cut-off

c. Branch or Product Determinations d. Payroll Period e. The 50 Per Cent Sales Test f. Lowest Established Rates g. Covered Employees h. Beginners or Probationary Employees i. Apprentices j. Other Items in the Wage Survey

B. List of Establishments in the Industry C. Effectiveness of the Panel Conference D. The Wage Survey

1. The Role of the Bureau of Labor Statistics

a. Confidentiality vs. Right of Cross Examination b. Validity of BLS Industry Lists

2. Survey Practice

E. The Hearing F. Tentative Decision

1. Definition (Including Product Divisions) 2. Locality 3. Prevailing Minimum Wages

a. Continued Use of the Median Technique b. Neglect of Lowest Established Rates

4. Beginners or Probationary Workers 5. Proposed Determination

G. Final Decision

IV. Judicial Review of Prevailing Wage Determinations A. The Lukens' Period: No Standing to Sue B. The Fulbright Amendment: More Apparent Than Real

1. The Death of Regional Determinations 2. Open Market Exception 8. Regional Determinations Revisited

C. Possible Alternatives V. Effect of Determinations

A. Available Data Limited B. Immediate Direct Impact of Walsh-Healey Determinations C. Adjustments of Wages in Higher Brackets D. Regional Impact E. Effect on Individual Plants

F. Effect on Future Hiring G. Other Effects

VI. What is Wrong-and Ways to Correct It

A. Review Within the Framework of the Act B. Procedural and Organizational Shortcomings

1. The General Problem 2. Proposed Remedies

a. The Panel b. Decision Making c. Review of Proceedings

C. Statistical Shortcomings

1. The General Problem 2. Proposed Remedies

a. Replace the Median Technique by One Which Gives Proper Consideration to Normal Industry Practice b. Use Lowest Established OR Lowest Paid Wages, De-

pending on the Circumstances in the Case

c. Make Proper Adjustments to Incentive Earnings

D. Conclusion

--- FOOTNOTES ---

INTRODUCTION EARLY in 1962, two groups of company presidents from different industries called on the Secretary of Labor in a single day to protest the adverse effect which they felt proposed wage determinations under the Walsh-Healey Public Contracts Act' would have on their respective industries. During the summer of 1962, a special

Committee on Walsh-Healey Act of the Advisory Council of Federal Reports completed a highly critical report to the Bureau of the Budget. The tone of that report in sum was that "the current approaches which are employed on an ad hoc basis by the Depart-

ment of Labor are often unfair, faulty, and/or unnecessarily burdensome on the particular industry involved."2

Almost simultaneously, the Federal Reserve Bank of Boston published a research report entitled "Effect of Walsh-Healey Minimum Wages on Regional Industries" which raised "the question of whether or not the Walsh-Healey Act has long outlived its usefulness."3 The Chamber of Commerce of the United States recently

0 S.J.D., University of Vienna, 1929; Management Consultant; President, Reports

Incorporated.

f Member, Bar of the District of Columbia; A.B., University of North Carolina, 1948; LL.B., Harvard University, 1951; Member of the firm Barco, Cook and Patton, Washington, D.C.

: Member, Massachusetts and District of Columbia Bars. A.B., Harvard, 1927, LL.B.,

1933. Member of the firm of Reilly and Wells, Washington, D.C.

See Table of Contents p. 266. 49 Stat. 2036 (1936), as amended, 41 U.S.C. §§ 35-45 (1958). ADVISORY COUNCIL ON FEDERAL REPORTS TO THE BUREAU OF TM BUDGET, REPORT OF

THE CoMTTrrEE ON WALSH-HEALEY Aar 2 (1962).

" BUEHNER, RESEARCH REPORT TO FEDERAL RESERVE BANK OF BOSTON, No. 19, EFFEC

OF WALSH-HEALEY MINIMUM WAGES ON REGIONAL INDUSnuEs 4 (1962).

[Vol. 1963: 20!

' CHAMBER OF COMMERCE OF THE U.S.A. PROCEEDINGS, WALSH-HEALEY BRIEFING CON-

FERENCE, Sept. 26, 1961.

5 CHAMBER OF COMMERCE OF THE U.S.A., WALSH-HEALEY & DAVIS-BACON BULLETIN,

Nov. 3, 1961.

eHearings on H.R. 9900 Before the House Committee on Ways and Means, 87th Cong., 2d Sess. 11-14 (Statement of Francis J. Trecker, President, National Mlachine Tool Builders' Association). See also Journal of Commerce, April 4, 1962, p. 4.

(1962)

7 52 Stat. 1060 (1938), as amended, 29 U.S.C. §§ 201-19 3 Recent industrywide proposed or final minimum wage determinations have been

1. Change in Approach

A second reason for serious concern is the announced "all-out" speed-up in determination procedures. In the past, the time lag from the initiation of proceeedings in an industry until the determination became effective is said to have averaged between 2 and 3

$1.43 for the metal business furniture industry, $1.55 for the office, computing and accounting machines industry, $1A8 for the fractional and $1.73 for the nonfractional horsepower motor and generator industry, and $1.80 for all but blueprint machine operators and draftsmen in the machine tool industry. Minimum hourly wage rates as high as $2.846 are in force for one industry in one district. * Proposed Labor Dep't Dec., 23 Fed. Reg. 2863, 2865

(1958) (drugs and medicine industry), refers to this as the "Congressional purpose." See also Proposed Labor Dep't Amend., 17 Fed. Reg. 11197 (1952) (textile industry).

,0 Reilly, Haslam, & Modley, The Threat of the Walsh-Healey Act, Harv. Bus. Rev., Jan. 1951, p. 93. See also Endicott Johnson Corp. v. Perkins, 317 U.S. 501, 507 (1943); Perkins v. Lukens Steel Co., 310 U.S. 113 (1940).

21 Brief for Appellants, p. 18, Consolidated Electric Lamp Co. v. Mitchell, 259 F.2d

189 (D.C. Cir. 1958), cert. denied, 359 U.S. 908 (1959).

1 48 DEP'T OF LABOR ANN. REP. 249 (1960).

[Vol. 1965: 205

"'Secretary Announces Improvements in Wage Determination Procedures, Dep't

of Labor News Release No. 4774, Sept. 18, 1961.

"4See, e.g., Tentative Labor Dep't Dec., 26 Fed. Reg. 10518

(office, computing and accounting machines industry), which asserts that "whenever the making of a determination has necessitated an answer to the question whether a determination should be grounded upon lowest established rates or the lowest rates actually paid reliance has been placed consistently upon the lowest rates actually paid."

BACKGROUND

25 See, e.g., Brief for the Lamp Industry Panel, Electric Lamp Industry, Feb. 15,

1962, pp. 13-16.

10 The legislative history given here is based largely on RADIO-ELECTRONICS-TELEViSON

M1ANUFACTURERS ASS'N, THE WALSH-HEALEY ACT: ANALYSIS AND SUGGESTED AMENDMENTS ch. 2, at 5 et seq. (1954) (RETMA is now Electronic Industries ASS'n); Note, Validity of Minimum Wage Determinations and a Consideration of the Need for the WalshHealey Act, 31 IND. L.J. 245 (1956); DERBER & YOUNG, LABOR AND THE NEW DEAL (1961).

17 Schechter Poultry Corp. v. United States, 295 U.S. 495 (1935). 18 SCHLESINGER, THE POLITICS OF UPHEAVAL 509 (1960). 19 See H.R. REP. No. 2946, 74th Cong., 2d Sess.

(1936); Hearings on H.R. 11554

Before the Subcommittee on Judiciary of the House Committee on the Judiciary, 74th Cong., 2d Sess. (1936); Hearings on S. 3055 Before the House Committee on the Judiciary, 74th Cong., 1st Sess. (1935).

[V/ol. 1963: 205

Steel Corp., 301 U.S. 1 (1937).

21 52 Stat. 1060 (1938), as amended, 29 U.S.C. §§ 201-19 (1958). 22 49 DEP'T oF LABOR ANN. REP. 246-75 (1961). 21 On Sept. 25, 1961, a minimum wage of $1.15 became effective for all industries subject to the act. Effective Date of Public Contracts Act Determination Changed to Sept. 25, Dep't of Labor News Release No. 4785, Sept. 25, 1961.

Vol. 1963:205]

SUMMARY OF DETERMINATIONS UNDER THE WALSH-HEALEY PUBLIC CONTRACTs ACT, MARCH, 196224

TOTAL

29 Summary of Minimum Wage Determinations Applicable Under the Walsh-Healey

Public Contracts Act, Dep't of Labor Releases, PC-14 (Rev. 11/61 and 2/62).

25All of these are in the bituminous coal industry. "049 Stat. 2036

(1936), as amended, 41 U.S.C. § 35

(1958); 41 G.F.R. §50-201.1

27 DEP'T OF LABOR, WAIsH-HEALEY PUBLIC CONTRACTS ACT, RULINGS AND INTERPRETA-

T'ONS No. 3, §§ 90-33, pp. 11-12, Jan. 31, 1961.

28 49 Stat. 2036 (1936), 41 U.S.C. § 35 (b) (1958). 29 U.S. CHAMBER OF COMMERCE PROCEEDINGS, WALSH-HEALEy BRIEFING CONFERENCE,

Sept. 26, 1961.

8o 80 CONG. REC. 10004 (1936).

3148 DEP'T OF LABOR ANN. REP. 249 (1960) (remarks of Secretary James P. Mitchell).

See also STRACKBEIN, THE PREVAILING MINIMUM WAGE STANDARD 2 (1939).

"STRACKBEIN, 35 23 Fed. Reg. 2865 (1958).

op. cit. supra note 31, at 110.

a' STRACKBEIN, op. cit. supra note 31, at 110. 85For a detailed discussion of the approach of the Public Contracts Board, see

SmAcKB iN, op. cit. supra note 31, at 109-34.

VCol. 1963: 205]

Act, N.Y.U. 14TH ANNUAL CONFERENCE ON LABOR 393, 417-19 (1961).

37 Modem Industry, Nov. 1950, p. 64. 11 26 Fed. Reg. 4174 (1961).

38 Id. at 92. See text p. 207 supra. "8 Ibid.

(Vol. 1963:205

40a Maverick determinations have been made for some industries in which a single In the company employed a very substantial percentage of all covered employees. tentative decision for the Photographic and Blueprinting Equipment and Supplief Industry, 25 Fed. Reg. 12522 (1960), the largest company employed about 50% of all covered employees. The Secretary determined as prevailing a wage which was between the minimum paid in the single plant of this company and the median of all othei establishments.

In the Electric Lamp Industry, Tentative Labor Dep't Dec., 28 Fed. Reg. 335C (1963), the largest company operated more than half of all establishments with more than one third of all covered employees. The Secretary determined as prevailing a wage which was between the median for the establishments including the largest company and the median for the establishments excluding the largest company.

"Tentative Labor Dep't Dec., 26 Fed. Reg. 10518

(office, computing, and

accounting machines industry).

12 49 Stat. 2036 (1936), 41 U.S.C. § 35 (b) (1958).

op. cit. supra note 31, at 44-45.

,SMACKBEIN, "Id. at 25. "Tentative Labor Dep't Dec., 27 Fed. Reg. 898 (1962) "'Od. at 901. '4 For a detailed discussion see Exceptions of the National Machine Tool Builders' Association to the Tentative Decision in the Determination of Prevailing Minimum Wages, Feb. 21, 1962, pp. 6-13.

(machine tool industry).

,9 STRACKBEIN, Op. cit. supra note 31, at 53. Al Id. at 49.

[Vol 1963,205

250 U.S. 1002 (1956).

1o Mitchell v. Covington Mills, Inc., 229 F.2d 506 51 STRACKBEIN, op. cit. supra note 31, at 31. 52 Mitchell v. Covington Mills, Inc., 229 F.2d 506 (D.C. Cir. 1955), cert. denied, 350

(D.C. Cir. 1955), cert. denied,

53 Proposed Labor Dep't Dec., 23 Fed. Reg. 2254 (1958) (paint, varnish, and related

U.S. 1002 (1956).

products).

5'229 F.2d 506, 508 5569 HARv. L. REv. 1341 (1956).

(D.C. Cir. 1955), cert. denied, 350 U.S. 1002

(1956).

The possible penalties under the act are: restitution of back wages; ten dollars per day liquidated damages for each day and each employee employed in violation of the child labor provisions; blacklisting of the violating contractor for three years; cancellation of the violator's contract with a requirement that it be let to another contractor and the violator be forced to pay any difference in cost to the government.05 In practice, the first two penalties are the only ones generally applied. Most violations found are adjusted in the field and never go to formal hearings.6

III

THE DETERMINATION PROCESS

According to the Department of Labor, industries are chosen for prevailing minimum wage proceedings on the basis of "importance of Government business, competitive advantages, the probability of prevailing minimum wage levels well above the present statutory minimum under the Fair Labor Standards Act, the significance of labor cost differentials in relation to contract bidding, and related factors."67 In some cases, proceedings are said to have begun after petitions from unions. The Rules of Practice say that "proceedings may be initiated by the Secretary upon his own motion or upon the request of any party showing a proper interest in the industry.68

Public Contracts Act, 57 were completed. During the same year, 3,638 safety and health inspections were made in industrial establishments performing on Government contracts under the Act. 49 DEP'T OF LABOR ANN. REP. 229-36 (1961).

05 RETMA, op. cit. supra note 16, at 39. "The budget of the Wage and Hour Division for 1963 is in excess of $17% million. Of this, $14 million was authorized for enforcement, $900,000 for wage determinations, and the rest for regulations and other purposes. GPO, THE BUDGET OF THE UNITED STATES GOVERNMENT FOR THE FISCAL YEAR ENDING JUNE 30, 1962, at 837.

e7 48 DEP'T OF LABOR ANN. REP. 250 (1960). 18 RETMA, op. cit. supra note 16, at 21. See also 41 C.F.R. § 50-203.15

eo ADVISORY COUNCIL ON FEDERAL REPORTS TO THE BUREAU OF THE BUDGET, REPORT

OF THE COMMITTEE ON WALSH-HEALEY Aar 3 (1962).

7 See, e.g., Tentative Labor Dep't Dec., 27 Fed. Reg. 1913

(motors and generators industry); Motors and Generators Industry Transcript, Oct. 3-6, 10, 1961, at 293.71

(1962)

REPORT or COIMITrEE ON BUREAU OF THE BUDGET, STANDARD

'WALSH-HEALEY ACT, op. cit. supra note 69, at 4. (1957). INDUSTRIAL CLASIFICATION MANUAL

7aId. at 431. 4 Thus, Major Group 35 comprises "Machinery, except electrical"; industry group 353 comprises "Construction, mining, and materials handling machinery and equip. ment"; and, finally, industry 3531 comprises "Construction machinery and equipment." Id. at 100.

75 See, e.g., Reilly, Haslam. & Modley, The Threat of the Walsh-Healey Act, Harv. (1961) Bus. Rev., Jan. 1951, p. 86; Tentative Labor Dep't Dec., 26 Fed. Reg. 10517 (office, computing and accounting machines industry), where "industry" comprised 80 per cent of a 3-digit industry group and from 85 to 100 per cent of two 4-digit industries within the 3-digit group.

76 See, e.g., Summary of Proceedings of Panel Conference for Construction Machinery and Equipment Industry, Feb. 20, 1962; Tentative Labor Dep't Determination, 25 Fed. Reg. 7801

(1960) (electron tubes and related products industry).

78 Transcript of Hearing, Motors and Generators Industry, Oct. 3-6, 10, 1961, In.

dustry Exhibit 15.

Vol. 1963: 20.5]

9 Summary of Proceedings of Panel Conference for Construction Machinery and

Equipment Industry, Feb. 20, 1962, at 2.

80 DEP'T or LABOR, WVAISI-HEALEY PUBLIC CoNTRAcTs Acr, RULINGS AND INTERPRErA-

TIONS NO. 3, pp. 12-13, Feb. 28, 1962 (emphasis supplied).

811 d. at 12, 14.

SBLS 2562, Earnings in the Manufacture of Surgical Instruments and Apparatus

(motors and generators industry); 27 Fed. Reg. 898

11 27 Fed. Reg. 1913

(1962) (1962) (machine tools industry). 8'27 Fed. Reg. at 3611-12. "See,

e.g., BLS 2760, Earnings in the Manufacture of Engines and Turbines (1961); BLS 2787, Earnings in the Manufacture of Construction Machinery and Equipment (1962).

8 A beginner, for purposes of the wage survey, is usually defined as a new employee hired at a rate lower than that established for a specific job during the period of time required to receive orientation or initial training for that job.

Beginner or probationary "tolerances" are usually granted only when there is a substantial number of establishments employing such employees. In the proposed decision for the flour and related products industry, 23 Fed. Reg. 5129 (1958), probationary rates were not granted because only 68 out of 480 establishments employed probationary employees and there were only 390 probationary employees.

In the tentative decision for the tires and related products industry, 24 Fed. Reg. 8741 (1959), the Secretary granted a beginner tolerance because 43 out of 64 establishments employing 43,406 out of 69,762 covered employees reported lowest established hiring rates. The Secretary said that in this case "the practice of employing beginners, however, appears to be sufficiently widespread . to provide separate rates for beginners. 24 Fed. Reg. at 8746.

."

.

[Vol. 1968:205

87 See, e.g., BLS 2699, Earnings in the Manufacture of Machine Tools (1960). 8 "Issuance of special certificates.

If, upon examination of the apprenticeship agreement, the Administrator or his authorized representative finds that the employment of the apprentice conforms to the requirements of the regulations in this part, he will issue a special certificate and mail one copy to the employer (who shall keep the same on file with his employment record) and one copy to the apprentice. The special certificate will authorize the employment of the named apprentice at the rate or rates less than the minimum wage applicable under § 6 and for the length or lengths of time specified in the apprenticeship agreement. Such rate or rates and the length of 29 C.FR. time for which they are applicable shall be set forth in the certificate." §521.5

(1949).

89 DEP'T OF LABOR, BuREAu OF EMPLOYMENT SECURITY, HANDBOOK ON

INDUSTR!AL

CLASSIFICATION 5 (1958).

that is likely to affect the ultimate determination. Industry's effectiveness in the panel conference is thus of critical importance. And yet industry's role in the conference is only consultative. The ultimate decisions as to what is going to be included or excluded in the wage survey rest with the Department of Labor.

Before the Department can "field" the wage survey, the proposed survey form (questionnaire) has to be approved by the Bureau of the Budget under its authority under the Federal Reports Act of 1942.90 This gives industry another opportunity to present its views either directly to representatives of the Bureau of the Budget or through the Advisory Council on Federal Reports, a group organized at the request of the Bureau of the Budget. We suggest that each industry subject to determination proceedings should be formally informed of its right to make representations before the Advisory Council and/or the Bureau of the Budget. Up to now, the scope of review by the Bureau and Council has been quite limited. The recent report of the Committee on Walsh-Healey Act of the Council gives hope that both the Council and the Bureau will become more effective.

The Office of Statistical Standards of the Bureau of the Budget currently asserts that its sole function is to make sure that questions in wage survey forms are clear, that the survey could be understood by those who have to respond, that the definitions employed are satisfactory for the purpose of the survey, and that adequate statistical procedures are employed.91 The Budget Bureau, on the other hand, presently believes that it has no authority with respect to what industries should be surveyed, and when. Nor would it have authority in regard to a Walsh-Healey Review Board92 or similar aspects of the administration of the act.

D. The Wage Survey

1. The Role of the Bureau of Labor Statistics

Once a survey questionnaire has received the approval of the Budget Bureau, the job of undertaking the wage survey is turned over by the Divisions to the Bureau of Labor Statistics (BLS). This Bureau is the government's principal fact-finding agency in the field

00 56 Stat. 1078-80 (1942), 5 U.S.C. §§ 139-139f (1958). 91 Minutes of Advisory Council on Federal Reports, Committee on Walsh.Healey

Surveys, Jan. 11, 1962, p. 3.

2d. at 10. 0

" GSA, UNrED STATES GOVERNMENT ORGANIZATION MANUAL 321 049 DEP'T op LABOR ANN. REP. 12 (1961). 05 BLS 2562, op. cit. supra note 82. 9 60 Stat. 237 (1946), as amended, 5 U.S.C. §§ 1001-11 (1958). 97 See, e.g., Industry Exceptions, Motors and Generators Industry, March 19, 1962,

(1962-63).

p. 16.

9 Summary of Proceedings of Panel Conference for Construction Machinery and

Equipment Industry, Feb. 20, 1962, p. 1.

" 118 F.2d 105 (6th Cir. 1941). 92 Baldor Electric Company v. Wirtz, Civil No. 3673-62, D.D.C., April 5, 1963.

100 Transcript of Hearing, Office, Computing, and Accounting Machines Industry,

April 26-29, 1960, pp. 79-89.

101 Labor Dep't Bulletin No. 1168, Techniques for Preparing Major Statistical

Series, Dec. 1954, p. 96.

EP'T OF LABOR, BLS, WAGE STUDmiES MANUAL OF PROCEDURES, Pts. II-IV, Nov.

1951, Sept. 1959, June 1958.

[Vol. 1963: 20,

..3 26 Fed. Reg. 10517-18 (1961). 10 Transcript, op. cit. supra note 100, at 81-82. 105 Ibid. 108 This and the following discussion are based largely on the WAGE STuDiEs

MANUAL OF PROCEDURES, op. cit. supra note 102, pt. II, ch. 1.

101 See BLS 2699, op. cit. supra note 87, where the tabulations prepared for a more

or less typical recent hearing are as follows:

Table 1. "Establishments and workers." This shows the estimated "universe" and

the actual number of establishments, total workers, and covered workers studied.

Table 2.

"Percentage distribution of covered workers by average straight-time hourly earnings." This table is prepared in every minimum wage case. While it was the critical table in the days of the application of the "cluster theory," its value and use at the present time is negligible.

Tables 3-5.

"Distribution of establishments and workers by lowest rate actually paid." The three tables show lowest wages paid to nonbeginners, beginners, and all covered workers. Under current practice, these are the crucial tables for the hearing. Under the median technique, the Secretary is likely to make a determination somewhere between the medians for "establishments" and "total covered worker employment" on these tables. Depending on the prevalence of beginner rates, the Secretary will either make a single determination based on the table including all covered workers or he will make a determination for nonbeginners based on the table showing nonbeginners and a "subminimum" determination for beginners, granting a "tolerance" for a limited time.

Tables 6-8.

"Distribution of establishments and covered workers by the lowest established job rate for nonbeginners, the lowest established hiring rate for beginners, and the lowest 'selected' rate for all covered workers." The three tables show the lowest established rates in force for beginners, nonbeginners, and all covered workers, regardless of what they are actually being paid during the payroll period surveyed. The questions leading to this table and the tables themselves have been omitted in recent questionnaires and tabulations.

Tables 9, 10. "Distribution of establishments and covered workers by the differential between hiring and lowest established job rate; and by time period required to progress from hiring rate to lowest established job rate." These two tables give information as to how much lower beginners' rates are than rates for experienced workers and for how long beginners' rates are in force (these tables were omitted in recent hearings). Table 11.

"Distribution of establishments and workers according to the number and per cent of workers earning less than specified amounts." This table, often called the "impact" table, shows how many, and what percentage of, workers, would be affected by a determination at a given figure. At present, this table is little used by the Department. Labor, however, has been using this tabulation to press the point that the wage data for plants in which one per cent or fewer of all covered workers earned less than the specified amount should be treated as if no worker earned less.

Tables 12, 13.

"Distribution of establishments and covered workers by amount of increase between payroll date and survey date, applicable to lowest rate actually paid." These tables (one including, the other excluding, beginners) tried to determine the extent of wage increases between the payroll period specified and the time the survey was undertaken. Note here the attempt to make a contractual wage increase "applicable" to lowest earnings, a statistically erroneous procedure which, however, received the approval of the Bureau of the Budget.

108 60 Stat. 258 (1946), 5 U.S.C. § 1003 (a) (1958). 109 See, e.g., 27 Fed. Reg. 3611

(1962) (scientific, industrial and laboratory instru-

ments industry); 26 Fed. Reg. 7550 (1961) (machine tools industry).

110 60 Stat. 244 (1946), 5 U.S.C. § 1010 (1958).

" Proposed Labor Dep't Dec., 23 Fed. Reg. 2863, 2866 (1958) (drugs and medicine industry); see also 23 Fed. Reg. 5354 (1958) (soap industry); 23 Fed. Reg. 2254 (1958) (paint, varnish and related products industry).

112Tentative Labor Dep't Dec., 27 Fed. Reg. 898, 901

industry). See also 26 Fed. Reg. 4173

(machine tools (1962) (electronic component parts industry).

118 Transcript, Machine Tools Industry, Sept. 12, 1961, p. 825. 214 Tentative Labor Dep't Dec., 27 Fed. Reg. 898, 901

(1962) (machine tools In-

dustry).

115 Transcript, Electronic Equipment Industry, Sept. 19, 1961, pp. 196-201. At this writing, there are understood to be legislative plans to include fringe benefits as "wages" in Walsh-Healey proceedings.

-327 Fed. Reg. 1913, 1914 12'Industry Brief, Machine Tools Industry, Oct. 18, 1961, pp. 55-58.

(1962).

lIndustry Brief, Motors and Generators Industry, Nov. 24, 1961, pp. 86-87. 2 Proposed Labor Dep't Dec., 23 Fed. Reg. 2863 (1958) (drugs and medicine in-

dustry), speaking of the "critical median level"; Tentative Labor Dep't Dec., 26 Fed. (office, computing, and accounting machines industry): "I Reg. 10517, 10518 regard both the median of the plants weighted by their covered worker employment and that of plants without such weighting as having significance in this proceeding." (electronic component parts industry) for extensive See also 26 Fed. Reg. 4173 (1961) discussion.

227 For a discussion of median see WAuxs & RoBEiT, STATrsncs-A NEw AppRoAcH

214-19 (1960); MORONEY, FAcrs FROM FIGURES 34-65

12a26 Fed. Reg. 10517, 10518

(1951).

(office, computing and accounting machines

industry).

220 26 Fed. Reg. at 10518-19 180 15 Fed. Reg. 3809, 3811 (1950).

(1961).

.

181 22 Fed. Reg. 3729, 3782 (1957). 232 ADVISORY COUNCiL ON FEDERAL REPORTS, REPORT OF THE PANEL ON PREVAILING MINIMUM WAGE SURVEYS UNDER THE WALSH-HEALEY PUBuc CoNTmAcrs Acr 6 (1950). The sentence quoted is taken verbatim from Letter from D. C. Phelps to Thomas Holland of the Dep't of Labor, July 26, 1939. This letter was concerned with the application of Walsh-Healey to the electrical manufacturing industry.

138 26 Fed. Reg. 5898, 5899 (1961). 18, Section 6 permits the Secretary to "make rules and regulations allowing reason-

able variations, tolerances, and exemptions to and from any or all provisions of [this (1936), as amended, 41 act] respecting minimum rates of pay ...... U.S.C. §40 (1958).

49 Stat. 2038

13r Industry Brief, Electric Lamp Industry, Feb. 15, 1962, pp. 41-48.

' 1321 Fed. Reg. 5444, 5445 (1956). 287 Final Labor Dep't Dec., 26 Fed. Reg. 4066 (1961)

(metal business furniture and

storage equipment industry).

CoPTRoLLER GENERAL, REPORT TO CONGRESS, REVIEW OF WAGE RATE DEtrRMNA-

TIONS FOR CONsTRUCTION OF CAPEHART HOUSING AT THE MARINE CORPS SCHOOLS, QUANTI. CO, VIRGINIA 58 (1962).

180 Dep't of Labor News Release No. 4774, op. cit. supra note 116. 140 41 C.F.R. § 50-203.22

(1962 rev.) provides that any minimum wage determination

[Vol 1963: 205

JUDICIAL REvIEw OF PREVAILING WAGE DETERMINATIONS

"shall take effect not less than 30 days after due notice is given of the issuance thereof by publication in the Federal Register, or at such time prior thereto as may be provided therein upon good cause found and published therewith."

11,1 Industry Exceptions, Machine Tools Industry, Feb. 21, 1962, pp. 27-81.

Vol. 1963: 205)

14s ELECTRONIC INDUSTRIES ASSOCIATION,

INDUSTRIAL RELATIONS DEP'T, RECOMMENDED

(1961).

REVISION, SAFETY AND HEALTH STANDARDS FOR FEDERAL SUPPLY CONTRACTS

"I, See 69 HARv. L. Rv. 1341 (1956). The economic, practical and legal wisdom f the safety regulations might well provide an illuminating subject for another irticle. A parallel study of the regulatory activities and enforcement techniques of :he President's Committee on Equal Employment Opportunity might also constitute an (1962 rev.).) These are all ippropriate subject for analysis. -elatively obscure but important aspects of the growth of the government's regulatory ?ower and influence which for various reasons have not been widely explored and held ap to the light of comparison. Except as they perhaps illustrate a trend, these are not, however, subjects which are within the scope of this article.

(See 41 C.F.R. § 60-1

1"5The act prohibits payment to employees of government contractors of wages less than the minimum wages as determined by the Secretary of Labor to be the prevailing minimum wages for persons employed on similar work or in the particular or similar .ndustries or groups of industries currently operating in the locality in which the -materials, supplies, articles, or equipment are to be manufactured or furnished under ;aid contract. 49 Stat. 2036 (1936), 41 U.S.C. § 35 (b) (1958).

(Vol. 1963:205

16 Lukens Steel Co. v. Perkins, 107 F.2d 627 (D.C. Cir. 1939), rev'd, 310 U.S. 113

(1940).

1,7 Id. at 630. 14s 310 U.S. 113 14 66 Stat. 308 (1952), 41 U.S.C. § 43a (b) (1958).

(1940).

50' 18 Fed. Reg. 471-72 (1953). 'In Covington Mills v. Mitchell, 129 F. Supp. 740 (D.D.C.) reed, 229 F.2d 506

(D.C. Cir. 1955), cert. denied, 350 U.S. 1002 (1956).

152 129 F. Supp. at 742.

18 229 F.2d 506, 508 n.2 (D.C. Cir. 1955), cert. denied, 350 U.S. 1002 (1956).

248 F.2d 635 (D.C. Cir. 1957).

18549 Stat. 2039 (1936), 41 U.S.C. § 43 (1958). 158 The court, however, brushed aside the Secretary's contention that judicial review of the open-market question could be had only in enforcement proceedings, and not in a proceeding to test the validity of the determination itself. It held that the mining concerns bringing suit were adversely affected or aggrieved within the meaning of the Fulbright Amendment, 66 Stat. 308 (1952), 41 U.S.C. § 43a (b) (1958), even though they had not entered into a contract to supply coal to the government. This case apparently overruled sub silentio the district court decision in Allendale Co. v. Mitchell, 12 Wage & Hour Cas. 760 (D.D.C. 1956), which had dismissed complaints of numerous manufacturers on the theory that standing to sue was not conferred upon manufacturers in an industry affected by a wage determination, unless they could show that they were parties to government contracts and paid less than the determined minimum.

17 259 F.2d 189 (D.C. Cir. 1958), cert. denied, 359 U.S. 908 (1959).

[Vol 1963:205

158 Judges Schweinhaut and Holtzoff in Covington Mills v. Mitchell, 129 F. Supp. 740 (DJD.C. 1955); Judge Tamm in Allendale Co. v. Mitchell, 12 Wage & Hour Cas. 110 (D.D.C. 1954).

139Judge Washington in Mitchell v. Covington Mills, 229 F.2d 506, 511 (dissenting opinion); Judges Vinson and Miller in Lukens Steel Co. v. Perkins, 107 F.2d 627 (D.C. Cir. 1939).

160 61 Stat. 146 (1947), 29 U.S.C. § 160 (f) (1958).

144 A.B.A.J. 1112 162 GPO, THE BuDGET OF THE UNrrED STATES GOVERNMENT FOR THE FISCAL YEARt

1962 at 837.

26s 52 Stat. 1060 (1938), as amended, 29 U.S.C. §§ 201-19 (1958). 161 Samuels, Effect of $1 Minimum Wage in Five Industries, 81 MONTHLY LAnoR

REv. 492 (1958).

"laHearings

Currency, Part 5, 82d Cong., 2d Sess. 2663, 2679 (1952).

on S. 2594 and S. 2645 Before the Senate Committee on Banking and

166 The percentages were computed by the authors on the basis of the "impact tables" adjusted for beginners but not adjusted for later corrections in the BLS tables. "17 Reilly, Haslam & Modley, The Threat of the Walsh-Healey Act, Harv. Bus. Rev.

Jan. 1951, p. 94.

68Industry Exceptions, Machine Tools Industry, Feb. 21, 1962, p. 19. 180 Reilly, Haslam & Modley, supra note 167.

['Vol 1963: 205

NATIONAL MACHINE TOOL BumsERS' Ass'N, MINIMUM

'WAGE DETERMINATON

THREATENS MAcHINE TOOL BUILDERS wITH 8.7 MILLION DOLLAR WAGE INCREASE, Preliminary Summary, Feb. 16, 1962.

17 See, for instance, New England Governors Committee on the Textile Industry and Senators Green, Lodge, Pastore, and Saltonstall as well as (then) Rep. John F. Kennedy in Hearings on S. 2594 and S. 2645, supra note 165, at 2531, 2742, 2746, 2790. 1 2 This should not be taken to imply that all small plants pay low minima. Quite a few are in the middle or higher ranges. But the plants at the bottom of minimum wage distributions are generally small.

facture of Machine Tools by the Lowest Rates Actually Paid, April, 1960

I's Dep't of Labor, BLS, Distribution of Establishments and Workers in the Manu(Table). 117 Dep't of Labor, BLS, Distribution of Establishments and Workers in the Manu-

facture of Electric Lamps by the Lowest Rates Actually Paid, June, 1961 (Table).

17 Dep't of Labor, BLS, Distribution of Establishments and Workers in the Manufacture of Motors and Generators by the Lowest Rates Actually Paid, Oct., 1960 (rev.) (Table).

17 Proposed Labor Dep't Dec., 23 Fed. Reg. 2863

(1958)

(drugs and medicine

industry).

DEPT OF COMMERCE, BUREAU OF THE CENSUS, THE PROPORTION OF SHIPMENTS OF

EACH INDUSTRY ACCOUNTED FOR THE LARGEST COMPANIES IN THE 1954 CENSUS OF MANUFACTURES

(1957).

18 E.g., the recent proceedings in the electric lamp industry.

279 Industry Exceptions, Machine Tools Industry, Feb. 21, 1962, pp. 14-17. o80See generally Reilly, Haslam & Modley, supyra note 167, at 94-96.

WHAT Is WRONG-AND WAYS TO CORRECT IT

leiTranscript of Hearing, Electronic Component Parts Industry, Mar. 31, 1960,

pp. 313-14.

111 Secretary Announces Improvements in Wage Determination Procedures, Dep't

of Labor News Release No. 4774, Sept. 18, 1961.

153 US. CHAMBER OF COMMERCE PROCEEDINGS, WALSH-.HEALEY BRIEFING CONFERENCE,

Sept. 26, 1961, pp. 13-14.

184 27 Fed. keg. 10163-64 (1962). 185 COMPTROLLER GENERAL, REPORT TO CONGRESS, REVIEW or WAGE RATE DETERMINA-

TIONS FOR CONSTRUCTION Or CAPEHART HOUSING AT THE MARINE CORPS SCHOOLS, QUANTICO, VIRGINIA 10, 11, 58, 59 (1962).

186 ADVISORY COUNCIL ON FEDERAL REPORTS, REPORT OF THE COMMI-TEE ON WALSH.

HEALEY PUBLIC CONTRACTS ACT (1950).

:187 ADVISORY COUNCIL ON FEDERAL REPORTS, REPORT OF THE COmirITEE ON WALSH-

HEALzY ACT, Appendix A (1962).

lee Dep't of Labor News Release No. 4774, op. cit. supra note 182. 180 See Modley in Aviation Week, July 3, 1950, p. 15. 200 See Bureau of the Budget, Circular No. A-40, Oct. 24, 1951, quoting Exec. Order No. 8248, 4 Fed. Reg. 3864 (1939); Exec. Order No. 10033, 14 Fed. Reg. 561 (1949); Exec. Order No. 10253, 16 Fed. Reg. 5605

(1951).

[Vol 1963: 205

of the top half of the industry. be standard practices, those of the bottom half "substandard."

This makes as little sense as calling the top half of all persons in an intelligence test "normal" and the other half "subnormal." This practice would make a genius with an IQ of 140 "normal," and a perfectly normal child with an IQ of 99 "subnormal." Yet, this is exactly the type of faulty reasoning which the Secretary now uses In the machine tool industry, for instance, the median plant in his determinations. paid $1.75 on a plant basis and $1.75 on an employment basis.

It is obvious that two small plants with 64 employees which paid minimum wages of $2.75 or more (a full dollar or more in excess of the median) are indeed exceptional It is equally obvious that and do not, in any way, represent the prevailing practice. (or a penny three establishments with more than 2,000 employees which paid $1.72 or so less than the median) are fully representative of standard industry practice. Yet, the Secretary's practice makes the exception a "fair and reasonable wage" and the standard "substandard."

Identification of the median with "prevailing" is wrong from the point of view of common sense. While the term "prevailing" is not used in statistical practice, statistical practice in comparable cases indicates different approaches as more rational.

1027Te controversy about what rate data should be used as the data underlying a determination has been summarized in the tentative determination for the electron tubes and related products industry, 25 Fed. Reg. 7801, 7802 (1960):

Management has emphasized, correctly, that the lowest rate actually paid in a particular plant may fluctuate from one period to another, according to whether any person has been hired in the lowest labor grade within the period required for an employee to complete his service at the lowest wage rate in that grade. Such a plant may have legitimate use for a lower wage rate for persons to be employed than it was actually paying at the time of the wage survey without reducing its existing wage structure. But even if the prevailing wages were determined from a table of lowest established wages, some plant would presumably find itself in precisely the situation to which the management group calls attention ....

The Secretary then went on to state "that some of the established wages which are lower than wages actually paid are not shown to have been paid sufficiently recently to give reasonable assurance that they will actually be used in future employments." The problem, then, is to make sure that a determination is based on the wage rates which plants "may have legitimate use for" but to exclude those for which there is no "reasonable assurance that they will actually be used in future employments." This, however, the Divisions fail to do.

The inflationary effect of the Secretary's position was clearly demonstrated in the recent lamp case. The survey of lowest wages actually paid in that industry was undertaken in June 1961, at the lowest time of employment. As employment increased after (See Industry Brief, Electhe survey date, lowest wages actually paid began to drop. tric Lamp Industry, Feb. 15, 1962, pp. 13-17.) The practice of the Secretary to rely upon lowest wages actually paid is statistically particularly objectionable because the Secretary adds contractually agreed-upon deferred rate increases to the de-facto wages paid at the time of the survey, disregarding the fact that at the time when these deferred increases take effect, the lowest wages actually paid may be substantially lower. In the electric lamp industry case just quoted, this erroneous practice led to errors as high as 14 cents.

(Id. at 24.)

193 The Secretary's handling of incentive earnings is also open to serious question from a statistical standpoint. The Secretary now requests that plants report total incentive-paid workers. including production bonuses for straight-time earnings Plainly, this practice flies in the face of the principle that the minimum wage of such workers is the guaranteed or base rate. This is the rate to which increments are added if the worker performs above the established minimum standard. If these increments are included in a determination, the new rate set is no longer a base rate. The value of the incentive system is thus likely to be destroyed.

It is to avoid this effect that the Army-Air Force Wage Board, which establishes pay schedules for several hundred thousand Army and Air Force employees, reduces hourly earned rates for incentive workers by 15 per cent, compares the reduced rate with the base rate, and uses the higher one of the two for determining prevailing rates. See, e.g., Kanninen, Rate Setting by the Army-Air Force Wage Board, 81 (1958). No comparable adjustment is made by the Monthly Labor Rev. 1107-09 Secretary of Labor.

104 Statistical practice in comparable cases indicates the need for proper consideration of dispersion around the median as one way to establish "normal" patterns. The same approach may be used to establish what is a "prevailing" practice.

If, for instance, we want to determine what intelligence quotients are "normal," we can start from the median and measure in both directions until we have a range of measurements which satisfies our objective. The number of measurements to be included depends on us; however, we must be sure that an equal number (or percentage) of measurements is taken above, and below, the median. If we should claim that 50 per cent is needed to qualify as "normal" or "prevailing" we would pick the 25 per cent of all persons who are directly below the median and the 25 per cent who are directly above. As a matter of fact, about half of all children in a test by

[Persons with an IQ (intelliTerman and Merrill in 1937 were rated as "normal." gence quotient) of 90 to 109 are classed as normal. See Woaw Boox ENCYCLOrEDIA 3806 (1947).]

(by establishments and by covered employment in these plants),

While it is not our purpose to argue that 50 per cent should be the range which makes a minimum wage "prevailing," the practice of measuring dispersion about the median, and equitably in both directions, is the only one that is statistically proper. This points to a determination technique which first establishes the medians of minithen mum wages establishes the appropriate ranges of the minimum wages to make them fit a reasonable concept of "prevailing," and finally determines a "prevailing minimum wage" between the two lowest points found in ranges for plants and for their covered employees. The fact that it is the lowest point in the range that must serve as point of determination is clearly recognized by the Secretary. In the proposed decision for the flour and related products industry, 23 Fed. Reg. 5129, 5132 (1958), he said: "[Tjhe lowest minimum wage paid in this half is the one most representative of the minimum wage practices.. and the best measure of the industry standard which I am directed " The reason why, in Walsh-Healey cases, a determination must be made to find .... 'not less at the bottom of the range is based on the act which requires payment of than" the prevailing minimum wages. (Industry Brief, Motors and Generators Industry, Nov. 24, 1961, p. 112.)

What coverage might make a range "prevailing" has been discussed in different (The American Paper and Pulp Association suggested a range industry presentations. that includes 95 per cent of the covered workers. See 1960 Hearing to Redetermine the Wages for the Paper and Pulp Industry.) A technique which has been called (For a detailed discussion, see Industry Brief, Motors the interquartile technique and Generators Industry, Nov. 24, 1961, pp. 107-12) extends this dispersion to the 25th. and 75th percentage points, the quartiles.

The argument against this approach in the tentative decision for the Electronic (1961 shows that the writer of the Component Parts Industry (26 Fed. Reg. 4173) tentative decision lacked comprehension of the statistical issues involved as well as of the economic facts.

We have seen that the minimum wages in the top half of wage distributions are

likely to represent primarily

a) exceptional situations in smaller plants, and b) the wage practices of the plants of a few major companies. The lowest rated jobs in such plants may differ substantially in job content, pro-

ductivity, and otherwise from those of plants in the lower half.

Even if sound statistical techniques did not require the combination of a measure of dispersion in case of use of the median, the fact that different jobs are likely to occur on the lower half of the distribution would necessitate consideration of this area. Only the very lowest part of the distribution should be excluded from consideration-just as much as the very highest part should.

Only equitable consideration of minimum wages below the median can, as a matter of fact, prevent serious damage to employers who are "paying fair and reasonable wages."

[Vol. 1965: 205