On this page:
All of the State unemployment compensation acts provide that benefits shall not be denied an otherwise eligible individual for refusing to accept new work "if the wages, hours, or other conditions of the work are substantially less favorable to the individual than those prevailing for similar work in the locality." This provision in the unemployment compensation acts is one of the most difficult to administer. Its application can best be understood in relation to the other benefit provisions in the State acts.
In order to be eligible for benefits under the State acts a claimant must meet the requirements of the law. Among other things he must be able to work and available for work; that is, he must be currently in the labor market. If he does not stand ready, willing, and able to accept suitable work during the week for which he has filed claim, he is ineligible for benefits.
In addition, though eligible, the worker may be subject to denial of benefits if his unemployment is due to a labor dispute, if he was discharged for misconduct connected with the work, or if he left his work voluntarily or has refused suitable work without good cause. Denial of benefits in such cases follows on the theory that the worker's unemployment is not due to a lack of suitable job opportunities.
These disqualifying provisions are in the nature of exceptions to the general remedial purpose of the acts. They deny benefits only if the claimant's action falls directly within the limits of the exception when all the facts and circumstances are considered. Under most State laws, for example, the claimant is subject to denial of benefits for refusing work only if the work was suitable and he refused it without good cause. Moreover, in determining whether the work was suitable for the claimant, most of the State acts specifically provide for consideration of the degree of risk involved to his health, safety, and morals; his physical fitness and prior training; his experience and prior earnings; the length of his unemployment and prospects of securing local work in his customary occupation; and the distance of the work from his residence.
The law does not specify the exact weight to be given these and any other considerations which may be relevant to the determination because whether a job is suitable for a particular worker and whether he had good cause for refusing it can only be determined on the basis of the acts in the case. Thus, the actual determination of whether a claimant is subject to disqualification for refusal of suitable work without good cause is left to the discretion of those charged with the administration of the act. The same is true of the availability provision and the other general disqualification provisions in the State acts.
As mandatory minimum standards, however, all of the State unemployment compensation laws in conformity with section 1603(a)(50) of the Internal Revenue Code, as amended, provide that an otherwise eligible individual shall not be denied benefits for refusing new work:
These requirements have been extended to all refusals of work in most of the State acts by providing that "notwithstanding any other provisions of this Act, no work shall be deemed suitable and benefits shall not be denied under this Act to any otherwise eligible individual for refusing to accept new work" unless it meets these three conditions. Clearly, "no work" is broader than "new work" and claimants are not subject to denial of benefits for refusing a job which does not meet any one of the three conditions under such a provision. Under some laws, the three labor standards requirements and the general criteria for determining whether work is suitable also apply to the determination of whether the claimant is subject to denial of benefits for voluntarily leaving work without good cause.
Inasmuch as the labor standards provisions are mandatory, they impose a duty on those administering the State act to assure themselves that the work offered meets these minimum standards before denying the claimant benefits for refusing work, regardless of whether he raises the issue. Inasmuch as they are minimum standards, they apply to all denials for refusal of offers or for referrals to new work regardless of his reasons for refusing the job.1 If the job is vacant as a direct result of a labor dispute it does not matter, for example, whether the claimant refused it on principle, because he was afraid of bodily harm in crossing the picket line, or because the employer wanted him to start work on Friday, the 13th. He is not subject to denial of benefits under the suitable work disqualification in any case. Neither may he be held ineligible for benefits because he is unwilling to accept work which does not meet these three minimum conditions. For example, a punch press operator who is unwilling to accept less than $.80 an hour may not be held ineligible for that reason if lower wages would be substantially less favorable than those prevailing in the locality for such work.
The labor standards provisions relate primarily to the conditions on the job as compared with conditions in like jobs and the manner in which they would affect the claimant. The availability and suitable work provisions, on the other hand, turn primarily on the nature of the work and the claimant's qualifications, circumstances, and prospects. Thus work which meets the labor standards provisions may not satisfy the suitable work criteria and may not be work which the claimant need stand ready to accept. For example, a job as a stenographer though it meets the labor standards requirement is not suitable for a file clerk who cannot type and take shorthand. Similarly, a job as a cook's helper which pays prevailing wages for such work is not suitable for an assistant chef who has been earning $60 a week and has prospects of earning as much again. Unless the work satisfies both the suitable work criteria and the labor standards requirements, the claimant is not subject to disqualification for refusing it and is not ineligible for benefits if he is available for a substantial amount of other work which is suitable for him.
If the three labor standards requirements, the first, which prevents denial of benefits for refusal of work if the job offered is vacant due directly to a labor dispute, was designed to preserve the neutrality of the State agency in labor disputes. The third, which prevents denial of benefits if the worker as a condition of being employed is required to join a company union or resign from or refrain from joining a bona fide labor organization, was designed to deter any effort to use unemployment compensation to impede or destroy labor organizations. The second, which prevents denial of benefits if the wages, hours, or other conditions are substantially less favorable to the individual than those prevailing for similar work in the locality, was designed to prevent the unemployment compensation system from exerting downward pressure on existing labor standards. It was not intended to increase wages or improve the conditions under which workers are employed, but to prevent any compulsion upon workers, through denial of benefits, to accept work under less favorable conditions than those generally to be obtained in the locality for such work.
It is with this second labor standard requirement that we are concerned in the succeeding discussion. The key words and phrases in this requirement are: "similar work," "locality," "prevailing," "substantially less favorable to the individual," and "wages, hours or other conditions of work." The interpretation given these phrases and the manner in which they are applied in each case determine whether the purpose intended will be achieved. Each of these words and phrases will be discussed in turn. Inasmuch as the requirement is intended to reflect labor market conditions, their interpretation should be based on existing labor market patterns and usage and they will be considered in that light.
Similarity of work can best be judged on the basis customarily used by employers and employees as a result of industrial experience: by occupation and grade of skill. As used in prior legislation, "similar work" has in fact been held to mean work in the same trade or occupation. Superficially this would seem to mean that a job is to be compared with others known by the same title.
However, job titles are sometimes misleading. Different occupation and grade designations are often used in different establishments for the same work. Conversely, the same titles are sometimes used for different kinds of work. The actual comparison of jobs must therefore be made on the basis of similarity of the work done without regard to title: this is, the similarity of the operations performed, the skill, ability, and knowledge required, and the responsibilities involved.
In some occupations the similarity of work cuts across industry lines and the differences in the manner in which the work is done are relatively minor. Bookkeepers and boiler operators, for example, are likely to do much the same kind of work whether employed by a grain elevator company, a manufacturing concern, or a retail clothing establishment. Either would be hired by establishments in almost any industry providing they had the necessary experience with the particular bookkeeping system or the heating plant in use and the required degree of skill. The essential similarity of work which cuts across industrial lines is generally true of most office, janitorial, and clerical occupations and to some degree of unskilled common labor.
In most occupations, on the other hand, there is likely to be considerable variation in the work done in different industries, in part of industries, or even in particular types of establishment within any industry. There are marked differences, for example, in the work of a glazier in the construction industry and one in the automobile or the furniture industry; and within the furniture industry between the work of a glazier on wooden furniture and one who works on metal furniture. Similar differences exist in the nature of the work done by a waiter in a "greasy spoon" and one in a hotel dining room and between the work of a dress saleswoman in a bargain basement and a salesperson in a dress salon. Thus even where there is an essential similarity, differences in the nature of the tools used, in the size and quality of the material worked on, or in the clientele to be served, may create characteristic differences in the work which are important to both employers and employees. Such differences are generally to be found in the mass-production-process and service occupations.
The nature of the services rendered may also be differentiated within an occupation category by the degree of skill and knowledge required. The work of a head bookkeeper in a large concern who sets up the bookkeeping system and assumes responsibility for it, is clearly different from that of a bookkeeper in charge of "accounts payable" or a posting clerk in the department. These differences are reflected in the wages and other conditions in their respective employments. The work of a regular salesperson who must have a thorough knowledge of the merchandise and who assumes responsibility for the stock is likewise to be distinguished from that of a rush-hour or counter clerk who is not required to have any specialized knowledge or who only accepts payment for articles selected by the customer.
The degree of distinction made within an occupation requiring the same basic skills depends to some extent on the degree to which the occupation is concentrated in the area. Where there is a heavy concentration, the workers become highly specialized and employers seek such specialization. As a result, minor differences in the work done are commonly recognized both on the job and in the hiring process.
On the other hand, the fact that "similar" makes allowance for some difference though it implies a marked resemblance must also be given weight. Too fine a distinction is likely to result in a comparison of identical rather than similar work. Generally, distinctions should be made within an occupation only when important differences in the performance of the job outweigh the essential similarity of the work.
In skilled trades a number of long-established and commonly recognized grades such as learners, apprentices, and journeymen will usually be found. There may also be special groups such as handicapped or superannuated workers which must be taken into account where there are actual differences in the tasks performed and speed and skill required. However, the work should not be distinguished on the basis of the kind of individual ordinarily hired for the job, since it is the work and not the worker which is to be compared under the law.
In conclusion, "similar work" is basically a common sense test. The degree of similarity required in any particular instance should be calculated to carry out the general purpose and spirit of the proviso. On the one hand the comparison should not be so broad as to result, for example, in the finding of a prevailing wage which bears no relation to those generally paid for some of the kinds of work being compared. On the other hand, the distinctions should not be so fine as to leave no basis for comparison with other work done in the locality and thus make meaningless the determination of the "conditions prevailing" for comparable work. Neither should the question of what is similar work be determined on the basis of other facts which are conditions of work within the meaning of the provisions, as for example, the hours of employment, the permanency of the work, unionization, or vacation, sickness, and retirement benefits. These other factors must be considered, but only after the question of what is similar work is decided. If they were considered in determining what is similar work, such consideration would beg the very question at issue: what conditions generally prevail for similar work?
The determination of what constitutes similar work is not difficult in occupations which have long been subject to union agreement. As a result of collective bargaining, the occupational duties and skill grades covered by agreement are usually well defined. Moreover, inasmuch as the definitions are based on industrial experience and the customs of the trade, they are applicable to nonunion as well as union in the locality.
In occupations and localities where the work in question has not been defined by mutual agreement between employers and employees, it is necessary to look to other sources. Guidance may also be derived from the job definitions and classification practices used by State and Federal agencies responsible for wage and hour data or the enforcement of minimum standards for various occupations, the employment service, employer groups, labor organizations, and the claimant's own experience. In the absence of such guidance a good general test of the similarity of the work is whether the duties and the skills required are sufficiently the same so that the workers employed in each of the jobs being compared could readily perform any of the others.
"Locality" like "similar work" is a somewhat indefinite term. Apart from any special reference to a particular place it means only a relatively limited geographic area. As used in the labor standards provisions it is an integral part of the concept of "the conditions prevailing for similar work." But while it is clear from the context that the conditions offered are to be compared with the conditions for similar work in the locality where the work is to be done, the nature and size of the area are not defined.
At first glance the use of arbitrary area limits such as city or county lines may appear persuasive because it seems easy to administer. Support for such interpretation is to be found in the public construction statutes in which the area for comparison of wages paid for similar work is generally defined as the State or civil division in which the work is to be performed. The phrase "immediate vicinity" in the Congressional Act of l862 governing the wage rates of unclassified navy yard employees has likewise been interpreted in terms of a 50-mile radius about the yard.
These definitions were adopted in large part to meet court objections to the use of so indefinite a term as "locality" where penal provisions are involved. This objection does not apply to the unemployment compensation laws nor is the same usage applicable. Unlike the public construction acts the unemployment compensation laws are not penal statutes. Unlike the Navy Yard Act, they do not deal with only one type of industry which is ordinarily concentrated in urban districts. Unemployment compensation agencies have occasion to deal with almost every kind of industry and with a variety of occupations, skilled and unskilled, organized and unorganized, which center in areas of varying size.
Defining "locality" by some arbitrary device such as city and county lines or a 50-mile radius about the establishment, without regard to the labor market pattern of the occupation, will in many instances fail to effect the intent of the prevailing conditions provisions. In some cases the area will be too large. In others, too small. If it is too large, it is likely to include more than one area of concentration for the same kind of work. In such cases, generalization of the conditions prevailing in several different areas of concentration is not likely to reflect the conditions actually to be obtained in any one of them. Similarly, if the limits are too narrow, the determination will reflect conditions prevailing in only part of the area in which those attached to the occupation ordinarily seek employment.
Results in better accord with the purpose of the labor standards provisions can be achieved by interpreting "locality" in terms of the area of immediate labor market competition for similar work. It is the variation in wages and other conditions in their customary occupation within the competitive labor market area in which they normally expect to obtain employment which immediately affects workers. Accordingly, "locality" as used in the labor standards provisions in the Internal Revenue Code and the State unemployment compensation acts may be defined as the competitive labor market area in which the conditions of work offered by an establishment affect the conditions offered for similar work by other establishments because they draw upon the same labor supply. The term "area" as used in section 103.50 of the Wisconsin statutes which provides that the hours of work on public highway projects shall be no longer than those prevailing for such work in the area is similarly defined as the locality from which labor for any project within such area would normally be secured. Definition of locality in terms of the competitive labor market area is also in accord with the practice of most unemployment compensation agencies insofar as can be discerned from the administrative decisions.
In establishing the competitive labor market locality for an occupation the dominant considerations are the location of the establishments employing similar services, the area from which (regardless of civil and political boundaries) workers are normally drawn to supply the needs of these establishments, the commuting practices and ease of transportation in the area, and the customary migration pattern of the workers in the occupation.
Because most industries tend to cluster in towns and cities, urban and metropolitan districts, including the suburbs and outlying area within ordinary commuting distance, generally constitute the locality for most industrial occupations. In some places two or three nearby communities with similar industrial activities may constitute a single locality for many occupations. Mill or mining communities in which the companies draw their employees from the surrounding territory in competition with each other are a good example. Similarly, heavy industrialized urban districts such as the San Francisco Bay area in which there are a number of communities within easy transportation distance of each other may constitute a single locality for occupations common to the entire area.
An extensive urban or metropolitan district may on the other hand encompass several localities for occupations in which the workers do not move freely from one community to another. The San Francisco Bay area, for example, apparently encompasses several different labor markets for domestic work in which different conditions may prevail because there is no direct competition for labor among employers or between those seeking such work in different communities. The same situation probably exists in other large urban districts such as the Chicago or New York Metropolitan areas and in many other fields of employment. To take an extreme example, the competitive labor market for pinboys in neighborhood bowling alleys may be no wider than several square city blocks. However, whether there is one or several labor market localities in an urban district for an occupation will vary from one place to another with the size of the district, the location of the establishments employing such services, the nature and ccustoms of the industry, and the commuting practices of the workers in the occupation.
The difference between determining the extent of the competitive labor market locality for similar work and determining whether the job a claimant was ordered is within reasonable travel distance from his home is discussed below under the heading "Distance to Work."
The competitive labor market for some kinds of work is not limited to urban districts and may encompass more extensive areas. In the logging occupations, for example, the entire lumbering region in which an offer of better wages by one of the operating companies at the beginning of a season would draw off workers from the other camps--or cause them to improve their conditions to meet the competition--would constitute the competitive labor market area. Similarly, the area in which structural steel workers or stone cutters ordinarily move from job to job and from which the contracting companies ordinarily recruit such workers may be regional or even Nation-wide.
Like variations are to be found in agricultural occupations. Thus, the immediate competitive labor market area for canning occupations would usually be more limited than that for field hands, while the customary migration pattern for the fruit and vegetable pickers involved will usually be more extensive. To follow the parallel further, while the competitive labor market area for poultry farm hands may be smaller than that for dairy hands in some places, the reverse may be true in other parts of the country where the poultry industry is more widespread and dairy farms are not clustered over large areas but scattered in small groups.
The size of the labor market locality should not be confused with the distance a claimant can reasonably be expected to travel to work. The first turns on the nature of the occupation and the economic character of the area. The second depends on where the claimant lives, his circumstances, and past work history. The two have little relation to each other. In large labor market areas, for example, the distance from one end to the other may be greater than a claimant can reasonably be expected to travel to and from work. Where the labor market area for the occupation is very small, on the other hand, it may be reasonable in view of transportation facilities to expect claimants to travel outside the labor market area. Some claimants may live far from the locality in which the job is offered. Some may have good cause for refusing jobs beyond the immediate vicinity of their homes. Others can reasonably be expected to commute a considerable distance in view of their past work histories and present circumstances. Regardless of the claimant's situation, however, the labor market locality in which offered work is compared with similar work to determine the conditions prevailing for the occupation remains the same.
There are no hard and fast rules for determining the locality for an occupation except that all of the factors which enter into the labor market pattern for such work should be considered in making the determination. A working knowledge of the nature of the occupation and the industries and kinds of establishments which employ such workers will usually be sufficient to indicate the relative size and general outline of the area. Information available from other agencies and groups which have occasion to deal with the same problems and the means to conduct a more complete study will also prove useful. In cases where the inclusion or exclusion of borderline districts or establishments would result in a substantially different determination, expert opinion and more thorough investigation may be necessary. Once the locality for the occupation has been determined, however, it can be applied in all future cases involving offers of similar work within the area, unless substantial changes in the industrial paattern of the area or the occupation become apparent.
While the prevailing standard was not applied to all conditions of work in earlier legislation, the standard has had long, and extensive statutory use. As applied to wage rates, its meaning was relatively well settled by administrative practice and court decisions prior to the enactment of the unemployment compensation laws. It may be assumed that those who framed the unemployment compensation acts were familiar with the legislative and court history of the standard. In the absence of evidence to the contrary, or of usage more appropriate to the intent of the provision, the standard in the unemployment compensation laws may therefore be construed on analogy to generally accepted usage under the prevailing wage provisions in prior legislation.
Under the earlier public construction statutes it has generally been accepted that the prevailing rate of wages means one specific rate for a given occupation in a given locality and not a number of rates all of which are prevailing. The prevailing, minimum wage requirement in the Walsh Healey Act of 1936, though it presents a somewhat different standard has likewise been interpreted to mean a single monetary figure in accordance with prior usage. It has also been generally accepted that "prevailing" means the most outstanding or commonly-paid rate, and that the prevailing, rate of wages for a given occupation and locality is a fact and its ascertainment a matter of investigation.
It may therefore be said as to each of different conditions of work to which the standard applies under the unemployment compensation acts: (1) that a specific condition of work is implied in each instance and not, for example, a range of wages or hours; (2) that the prevailing condition is that which most commonly obtains in the locality for similar work; and (3) that the determination of the prevailing condition is a matter of investigation.
From time to time there has been some question as to whether the prevailing standard in the unemployment compensation acts is to be applied in terms of the conditions under which the largest number of workers are employed or in terms of the conditions offered by the greatest number of employees. In some instances the conditions of work offered by the greatest number of employers has apparently been used because the information could more readily be obtained in that form. Where all the establishments involved are about the same size so that the greatest number of workers in the occupation are necessarily employed by the greatest number of employers, the result is much the same whichever test is used, if all the workers in the same establishment are employed under the same conditions. However, where the establishments are not the same size or the conditions within the establishments vary, the results are likely to differ widely depending on whether the test used is the conditions under which the largest number of workers are employed or the conditions offered by the greatest number of employers.
This issue has not apparently arisen under other laws. Under the public construction statutes for example, the prevailing standard has customarily been applied in terms of the rate paid the largest number of workers. Justification for this usage under the unemployment compensation acts is also to be found in the traditional use of the terms "prevailing wages" and "prevailing conditions of work" by economists and other social scientists as meaning the wages and other conditions under which the largest number of workers are employed. The chief merit of using the largest number of workers lies however, in the fact that it sets up the standard most consonant with the purpose of the prevailing conditions of work provisions. This can best be illustrated in terms of wages since that is generally the most important factor in the employment relation.
The upward or downward pressure which an employer exercises on the conditions offered for similar work in the competitive labor market locality is directly related to the number of workers he employs. An offer of better wages by a large establishment which employs several hundred welders will draw such workers from almost every establishment in the locality which pays less. Moreover, it will force employers who pay less to increase their wages if they wish to retain their employees and attract new workers. A similar increase in the wages offered by a shop which employs two or three welders will have little if any effect on the general level of wages in the occupation. Conversely, a cut in wages by a large establishment is likely to result in a reduction in the wages paid by other employers, while a similar decrease by a single small employer will have little effect on existing rates.
In other words, it is not the number of employers or how many different rates are paid but the number of jobs at each rate and level of wages which directly affects the individual worker's position in the labor market. By establishing the prevailing wage on the basis of the amount paid the largest number of workers, existing conditions in the labor market are, therefore, more truly reflected. Moreover, because each rate is weighted in proportion to the number of workers employed at that rate, the cumulative effect of the wages paid by numerous small employers is balanced against the wages paid by larger establishments.
As a general rule it may therefore be said that the prevailing wages, hours, and other conditions of work are those under which the largest number of workers engaged in similar work in the locality are employed.
Under the public construction acts, the rate paid a larger number of workers than any other--that is the most common or modal rate--has generally been recognized as that prevailing, where a majority of the workers in the occupation are employed at the same rate. The mode is also generally used where less than a majority, but as much as 30 percent or 40 percent of the workers are paid at the same rate.
In the event that less than 30 percent or 40 percent are paid at the same rate, the average of all the rates paid weighted by the number of workers at each rate** is generally used rather than the mode. The New York Public Construction Act, for example, provides that the average shall be used if less than 40 percent of the workers in the occupation are paid at the same rate. Under the Federal Davis-Bacon Act, the average is used if less than 30 percent are paid at the same rate.
As applied to wages and hours and such other conditions as can be measured in numbers, a combination formula of this kind best carries out the intent of the prevailing conditions of work provisions to prevent denial of benefits for refusal of work if the conditions are substantially less favorable than those generally to be obtained in the locality for similar work. This follows because each of the two methods, the mode and the average, is used under the circumstances to which it is most applicable.
The indented material below provides a more complete explanation of the methods of determining the prevailing condition of work. It may be skipped by those interested in the broader aspects of the subject.
The mode is used so long as one condition of work clearly prevails over all others and is therefore most representative of those to be obtained in the locality. This method has the merit of utilizing a condition of work which actually exists as the standard. It also has the advantage of being relatively easy to use because it requires no calculation beyond ascertaining which of the existing conditions is most widespread.
The average, on the other hand, is used where the largest number of workers employed at the same wages or hours or other condition of work does not constitute a substantial proportion of the total number in the occupation. Where this occurs, the condition under which the largest number of workers are employed in the occupation may not always be representative of those generally to be obtained. In such cases results in better accord with the purpose of the prevailing conditions of work provisions can usually be achieved by using the weighted average. In the case of wages, for example, this method, because it reflects the entire range of wages and the number of workers employed at each level of earnings, usually yields a wage which is more representative of those generally to be obtained in the locality than that paid any relatively small proportion of the workers in the occupation.
However, since conditions like seniority rights, which cannot be measured in numbers, cannot be averaged, the mode must of necessity be used in determining the prevailing condition of work where such factors are involved, even though only a small percentage of the workers in the occupation are employed under the same condition. The mode also should be used in determining the wages or hours prevailing for similar work even though there may be relatively few employed under the same condition, if the information necessary to calculate the average is not available. Conversely, where the average is known, but the information necessary to obtain the mode cannot be obtained, it may be necessary to use the average wage or the average number of hours as the standard for comparison even though a substantial number of workers may be employed at the same wages or hours.
Use of Class Intervals.--In determining the mode it is often simpler to divide the entire range of wages or hours or other conditions existent in the locality into class intervals rather than count the number of workers employed under each particular condition. For example, the number of workers employed at different wage rates may be ascertained on the basis of 2-cent or 5-cent or 10-cent class intervals depending on how great the amounts involved are. That is, the number of workers employed at different rates may be reported in terms of the number, receiving 60 to 64.9 cents an hour, the number receiving 65 to 69.9 cents an hour, and so forth rather than the number receiving 60 cents an hour, the number receiving 60.5 cents an hour, the number receiving, 61 cents an hour, and so on. If the information is received in this form and the actual mode is not known (1) the modal point in the most numerous class may be determined through the use of one of the statistical formulas designed for that purpose, or (2) the mid-point of the most numerous class may be used with due allowance for the fact that it is only an approximation.
The weighted average may also be derived on the basis of class intervals (1) by multiplying the mid-point of each class interval by the number in the class, adding the totals, and dividing the result by the total number of workers involved or (2) by using one of the shorter statistical formulas designed for the purpose.
Ordinarily the factual information needed to ascertain the conditions prevailing in the locality for similar work can be obtained from labor and employer organizations, from representative employers and employees, from the Employment Service, or from other Government agencies which are responsible for the collection of data on ages and hours, the enforcement of minimum labor standards in various occupations, or the administration of industrial safety codes and the like. If conditions in the occupation are fairly stable, information once obtained may prove useful over a considerable period. This is particularly true in the case of occupational wage rates which, in normal times, are likely to remain unchanged over long periods. It may therefore prove useful to construct tables of occupational rates and keep them on hand for ready reference. These should be amended from time to time as better or more current information becomes available.
The determination of the conditions prevailing in the locality for similar work is comparatively simple where most of the workers in the occupation are employed under uniform collective bargaining agreements or where the conditions are governed by custom or law. More extensive investigation and more careful examination of the data available is usually required where there are relatively few workers employed at the same wages or hours or other conditions of work. Even in such cases, though, sufficient information can generally be obtained to enable a reasonably accurate approximation.
Thus where only the range of wages or hours is known a point nearer the middle than the bottom of the range may be used as a rough estimate since there are normally few workers at either extreme. If there is reason to believe that a larger number than usual are nearer the top or the bottom of the range the estimate may be moved up or down accordingly.
Similarly, where the most complete and accurate information available is not entirely current, allowance may need to be made for any noticeable upward or downward trend which may have taken place in the meantime. In other instances in which accurate information of the conditions under which such workers are currently employed in the locality is lacking, typical offers made through the Employment Service or other channels may provide some guidance. The claimant, if he is familiar with the conditions which generally obtain for such work in the particular labor market locality, may also be able to provide some information.
In each case, though, it is for the unemployment compensation agency or tribunal to sift the data and to make the determination on the basis of the best information available.
Many of the conditions of work to which the prevailing standard is applied under the unemployment compensation acts, like seniority and safety provisions, do not lend themselves to exact comparison. In considering factors of this kind it cannot always be determined whether one condition or combination of conditions is less favorable than another. Even in the case of wages and hours which can be more exactly compared, the wages or hours which in fact prevail cannot always be definitely determined. Nor can the conditions of a job in question always be foretold with certainty. The rate of earnings, for example, will in many instances depend on the individual's ability. Working hours may also be subject to variation under different circumstances so that even the employer cannot say exactly what they will be. Moreover, a condition which is important in one occupation and locality may be relatively unimportant in another. For example, the use of ventilators to draw off fumes is important in a chemical plaant and the height of a chamber to which he is assigned may be important to a miner. Both are relatively unimportant, however, in office work.
A certain amount of leeway has therefore been allowed in the application of the prevailing standard under the unemployment compensation acts by providing that benefits shall not be denied otherwise eligible individuals for refusing work if the wages, hours, or other conditions are substantially less favorable to the individual than those prevailing.
The provision thus presents a definite but not an inflexible standard. It does not preclude the denial of benefits for refusal of work where only minor or purely technical differences are involved which could neither undermine existing labor market standards nor have any appreciably adverse effect on the worker. It also allows a reasonable margin for error where the conditions prevailing in the locality for similar work or the corresponding conditions of the work offered cannot be exactly ascertained. But the basis of comparison in each instance, insofar as they can be determined, is still the conditions under which the greatest number of workers in the occupation are employed in the locality.
The meaning of the words "not substantially less favorable to the individual" cannot be defined in terms of any fixed percentage, amount, or degree of difference. Both the actual condition in question and the extent of the difference, as well as its effect on the worker, must be considered in each case.
If the conditions of the work the claimant refused and those prevailing are known, it is usually easy to determine whether the difference is of a material or substantial nature or is of no real consequence. In borderline cases where it is not clear whether the difference is material, the general rule that remedial legislation is to be liberally interpreted and applied in favor of those it was intended to aid would indicate that the claimant be given the benefit of the doubt. Similarly, when the facts cannot be precisely determined, the claimant would not be subject to denial of benefits for refusing work unless it is reasonably certain that the conditions on the job are not substantially less favorable than those prevailing.
There are some situations in which the prevailing standard provisions are not directly applicable though the work is unsuitable because the conditions of employment are substandard. Thus, though the conditions prevailing for similar work in the locality will ordinarily be better than the minimum standards set by State or Federal law, investigation may occasionally reveal that the wages, hours, or other conditions prevailing in a particular occupation and locality are below the applicable legal minimum. In such cases where the conditions of the work offered are in violation of law, even though they are not substantially less favorable than those prevailing, the claimant has good cause for refusing the job under the general suitable work provisions in the State acts. It is well settled that one law should not be so applied as to cause or result in the violation of another.***
Similarly, the claimant generally has good cause for refusing a job if the wages or other conditions are far less favorable than those in most other kinds of work in the locality, for which he is qualified, even though the job or the work in question is not covered by State or Federal wage and hour legislation. In view of the wages and other conditions generally to be obtained in the locality in other employment which the claimant is able to perform, such work would ordinarily be unsuitable and the claimant would have good cause for refusing it under most State acts. Payment of benefits in cases of this kind is also in accord with the intent of the prevailing conditions of work provisions to prevent operation of the unemployment compensation acts to depress the general level of working conditions through denial of benefits for refusal of substandard employment, though they may not come squarely within the letter of the provisions.
*This is a reprint of the statement "Principles Underlying the Prevailing Conditions of Work standard," originally issued by the Bureau of Employment Security as Unemployment Compensation Program Letter No. 130, dated January 6, 1947.
**i.e., each rate is multiplied by the number of workers employed at that rate, and the sum of the totals is then divided by the total number employed in the occupation to obtain the average rate.
***From another point if view it might also be held (1) that the conditions "prevailing" for similar work means legally prevailing, (2) that only conditions of work which meet the applicable State and Federal statutory standards should be considered in determining the conditions prevailing for similar work, and (3) that conditions which violate statutory standars do not meet the requirements of the prevailing conditions of work provisions. Under such an interpretation, the prevailing conditions of work provisionswould also prevent denial of benefits to claimants who refused work under conditions which were in violation of the law.
In the public construction acts the prevailing standard has generally been applied in terms of the prevailing "rate of wages" or the prevailing "rate of per diem wages." It has been argued that the word "wages" as used in the prevailing conditions of work provisions in the unemployment compensation Act also means the wage rate.
Support for this view is found in the fact that the hours of work, which in conjunction with the wage rate largely determine the earnings of most workers, are specifically set forth as a separate consideration. Accordingly, the provisions that benefits shall not be denied for refusal of work if the wages are substantially less favorable than those prevailing have at times been taken to mean that the hourly wage rate may not be substantially less than that prevailing.
This usage may be appropriate for the purpose of establishing the minimum rate which may be paid workers in various occupations under government supply and construction contracts. However, it is not the purpose of the prevailing conditions of work provisions in the unemployment compensation acts to establish a minimum rate which may be paid, but to prevent downward pressure on existing conditions and to give the claimant the benefit of conditions which are not substantially less favorable to him than those prevailing in the locality for similar work. Comparison in terms of wage rates alone is not always sufficient to accomplish this purpose.
Earnings are frequently affected not only by the wage rate and the hours of work, but also by the method of payment, the overtime practices, and various extra bonuses and premiums. For this reason, workers generally look to both the rate and the total weekly earnings in determining whether they will accept a particular job or continue to seek other work. Similarly, employers do not merely announce the rate of pay but also emphasize total earnings. In addition, all methods of payment do not lend themselves to comparison in terms of wage rate. Though most workers are now paid at hourly or piece rates, some are still paid a flat daily or weekly wage regardless of the hours put in or the amount of work done. It is only by taking all of the factors which would affect the claimant's earnings and those of most workers in similar employment in the locality into consideration that it can be determined whether the wages offered are less favorable than those prevailing.
Thus, where most of the workers in a particular occupation and locality are not paid on the basis of the amount of production or sales completed or the hours of work put in, but are paid a monthly or yearly salary, as is frequently true in the case of managerial and professional employees as well as farm hands, the wage comparison must be made in terms of their total monthly or yearly earnings including any remuneration received in addition to the base salary. Similarly, if the hours in the occupation are irregular and most of the workers are paid at hourly or piece rates or on a percentage basis as in the case of longshoremen, home workers, and many taxicab drivers, the comparison must be made in terms of hourly or piece rates or on a percentage basis. In such cases, the fact that the hours are irregular and unscheduled prevents any further comparison of earnings.
However, in the great majority of occupations in which the workers are paid fixed or variable rates or commissions, so that their earnings depend in large part on the actual hours of work, both the wage rates and the weekly wages can be compared and both need to be taken into consideration to determine whether the wages offered are less favorable than those prevailing.
Where some of the workers are paid at other than time rates or receive variable incentive wages in addition to the hourly base rate, the various rates may be compared in terms of average straight time hourly earnings. In such cases, the average straight time hourly earnings may be derived by dividing the weekly wage minus overtime earnings by the weekly hours of work less the overtime hours. If other nonproduction bonuses or premiums are paid in addition to overtime, these would also have to be subtracted from the weekly wage before dividing.
Conversely, where the weekly wages are not directly comparable because of differences in the hours of work, the prevailing weekly wage may be derived by multiplying the prevailing hourly earnings by the prevailing hours of work. If the hours usually include overtime, the overtime earnings would also have to be taken into account in determining the prevailing weekly wage. For this purpose prevailing overtime earnings may be estimated on the basis of the usual overtime rates and practices in the occupation and locality. Any other nonproduction premiums or bonuses customarily paid workers in the occupation would likewise have to be taken into consideration in such cases in determining the prevailing weekly wage.
Implicit in the comparison of both the hourly rate and the weekly wages is the general rule that the wages offered will ordinarily be substantially less favorable to the worker than those commonly to be obtained in the locality for similar work if either the hourly or weekly earnings are substantially lower than those prevailing. If, for example, the work in question is usually done on a full-time basis, the wages entailed in an offer of part-time work would usually be substantially less than those of most workers in similar employment even if the hourly rates were the same. The wages he would earn in part-time employment would therefore be substantially less favorable than those prevailing in the occupation for a worker who is seeking full-time work. Similarly, if the hourly rate were substantially less than that prevailing, the wages would generally be substantially less favorable than those of most workers in similar employment. This would hold true even though the job paid higher weekly wages thaan most such jobs because the hours of work were longer. In such cases, the conditions of the work offered would be substantially less favorable than those prevailing both because the hourly rate was lower and the weekly hours were longer than those generally to be obtained. The claimant would not therefore be subject to denial of benefits whether either or both factors were taken into account.
In some cases, however, a true comparison may require further analysis. Other factors that affect the weekly and hourly wage may also have to be taken into consideration. Thus the payment of overtime or other nonproduction premiums and bonuses over and above those ordinarily paid such workers in the locality may have a bearing on whether the hourly rate of earnings is actually less favorable than that prevailing. To illustrate: most of the workers in the occupation may be paid at straight time rates with nothing additional for overtime, and the prevailing hourly rate may be $.70 an hour, the prevailing weekly hours of work 48 and the prevailing weekly wage $33.60. The job in question on the other hand, may pay only $.65 an hour. At straight time rates, this would amount to only $31.20 for a 48-hour week and would be substantially less favorable than the wages prevailing for similar work in the locality. However, the wages may not be less favorable if other factors enter the picture. If, for example, the job paid time and a half after 40 hours, the worker would earn $33.00, which is somewhat more than the prevailing wage for the same work week. In effect, he would be earning a bit more than the prevailing rate of $.70 an hour.
In other instances, the provision of special benefits over and above those received by most workers in similar employment in the locality may make the wages as favorable as those prevailing. Thus the fact that the worker would be paid for vacation and sick leave has been taken into consideration in determining whether the wages were substantially less favorable than those of most workers in the occupation. It should be remembered, however, that such benefits may not outweigh the difference in the money wages the worker would earn the year around. In addition, while workers may appreciate benefits of this kind if they are afforded in addition to the usual wage, they may prefer to receive the difference between the wages paid and the usual wages for such work in money rather than in other forms because of the greater freedom it gives them to purchase the goods, leisure, or services they want.
The question of differential payments for evening or night work in the form of equal pay for shorter hours or a higher rate or additional bonus may also arise. If such differentials are ordinarily paid they need to be taken into account. Accordingly, a claimant who refuses employment on the night shift at the wages which are ordinarily paid for day work but which are substantially less favorable than those prevailing for night work, would not be subject to denial of benefits under the prevailing conditions of work provision. A like result would be reached where there were established differentials for jobs involving special risks to health or safety beyond those ordinarily incurred in the occupation, as in the case of mine operations carried on in water. In cases of this kind, there may also be some question as to whether the work is similar to the less dangerous or easier operations with which it is being compared. But the same result as to payment or denial of benefits should be reached whether the jobs are held to be different with different wages prevailing for each, or whether the work is considered similar and the practice of paying a differential rate is taken into account.
In some occupations it may also be necessary to allow for temporary differences or seasonal fluctuations in hourly and weekly earnings both in determining the prevailing wage and in determining whether the wages offered are substantially less favorable than those of most workers in similar employment. It is ordinarily expected, for example, that the earnings of department store sales workers who are paid a commission in addition to their hourly rate, will reach a peak during the winter holidays and be relatively low during the summer lull. Similar variations are to be found in the garment trades and in many other occupations in which the hours of work and consequently the weekly earnings are reduced during the off season. Since all of the establishments involved will not be affected simultaneously or to the same extent it is best to determine the prevailing wage in such cases on the basis of a normal period whenever possible, and to compare the wages offered with those prevailing in terms of the normaal earnings of other workers in the establishment. If the experience of other workers in similar employment offered in the establishment indicates that the earnings in the job will average as much as those of most workers in the occupation and that the fluctuations will be no more frequent and no greater than is ordinarily to be expected in such employment in the locality, due allowance may be made for such differences. If, however, the wages do not average as much as those of most workers or the fluctuations are so extreme as to render the earnings even more uncertain than those of most such workers, the conditions of the work offered may be substantially less favorable than those generally to be obtained for similar work.
A somewhat different problem is presented where most of the workers in the occupation are paid on the basis of progressive wage scales such as are frequently used by large establishments and incorporated in union agreements. In certain industries and plants, for example, inexperienced workers are hired at a minimum entrance rate and their wages increased during the training period until they are receiving as much as other workers in the department. Experienced workers may likewise be hired at a minimum job rate and their wages gradually increased up to the maximum rate paid by the plant for such work. In some cases the increases may be based on length of service with the employer; in some cases, on merit (i. e., usually skill and experience and speed); in others, on a combination of both.
Where progressive wage scales prevail, workers cannot ordinarily expect to be hired at the wages currently being paid the greater number currently employed in the occupation because many of those employed have received periodic increases based on the length of time they have worked in the same establishment. Accordingly, where progressive wage scales prevail, the determination of whether the wages offered are substandard is generally made not on the basis of the prevailing wage but on the basis of the prevailing wage scale. Determination of the prevailing wage scale involves consideration of at least three factors: (1) the prevailing entrance rate: (2) the basis on which the rates are increased: and (3) the amount and frequency of the increases. The need for considering all three of these factors when applying the prevailing wage standard where progressive scales are involved can readily be illustrated.
One illustration may be found where the rate increases in a particular occupation and locality are based on length of service alone, and new employees are almost invariably hired at the entrance rate. In such cases an offer of work at the prevailing entrance rate for inexperienced workers, or the prevailing minimum job rate for experienced workers, would not ordinarily be considered substandard inasmuch as most of the workers in the occupation are hired on the same basis and at the same rate. Nevertheless the wage scale offered may still be substantially less favorable to the worker. For example, if the greater number of workers in the occupation are hired at $.70 an hour and move up to $1.10 within a year, an offer of $.75 with increases up to a maximum of only $.90 after a year on the job would be substantially less favorable than the prevailing scale of rates.
On the other hand, where workers are not always hired at the entrance rate, and rate increases depend at least in part on skill and experience, it may be that a worker with prior experience in the occupation can expect to be hired at more than the entrance rate. In such cases an offer of work at the minimum rate might well be substantially less favorable than that prevailing for a worker who has formerly earned a rate above the minimum or the middle of the range. Investigation will usually reveal the customary hiring practice in regard to workers with varying degrees of prior experience and skill and whether the entrance rate and the rate scale are as favorable to the claimant as those prevailing.
Aside from its effect on the amount the worker earns, the method of wage payment is itself an important condition of work. Workers frequently have justified objections to employment under a different method of payment than that to which they are accustomed and long and bitter strikes have been fought over changes from time work to piece work and the introduction of incentive wage systems. Even though the wages offered equal those of most workers in similar employment, it may therefore be necessary to determine whether the method of payment is substantially less favorable than that prevailing.
As a condition of work, the method of wage payment may be substantially less favorable to the worker than that prevailing: (1) if it would yield substantially lower earnings than the prevailing method; (2) if the earnings would be more irregular or less certain than under the prevailing method; or (3) if it would require the worker to work faster or under greater tension than the prevailing method of payment. Generally, however, the customary practice of the trade in the locality in which the work offered will govern the decision as to whether a system of payment found objectionable by workers is substantially less favorable than that prevailing.
In occupations in which the hours are not scheduled by the employer, either directly or indirectly, they are not a condition of the work and do not enter into consideration in determining whether any of the conditions of the work offered are substantially less favorable than those prevailing in the locality for similar work. Where the hours are regulated by the employer, they are second in importance only to wages. Together with the wage rate and the method of payment they largely determine the worker's earnings. In themselves, they determine the title the worker must put in on the job and the time he has for his own needs and leisure.
Aside from their effect on the worker's earnings, the hours of the work offered may be substantially less favorable than those prevailing, in the locality for similar work, if they are substantially longer, or less convenient. If "wages" as used in the prevailing conditions of work provisions is deemed to mean only wage rates and not weekly wages, it may also be held that substantially shorter hours than those prevailing which would result in lower earnings, are substantially less favorable to a claimant who is seeking full-time employment.
Inasmuch as most workers are employed at regular hours which are limited by industrial practice and custom, it is not usually difficult to ascertain the hours prevailing in the locality for similar work and to determine whether the hours of the work offered are substantially longer than those prevailing. Generally it is not necessary to consider the possibility of extra overtime in making the determination. If, however, a considerable amount of extra time beyond the regular weekly schedule is frequently required of workers in the occupation Or the evidence indicates that it would be required on the job in question, that would also have to be taken into account. In such cases the past experience of other workers in the establishment may offer some guidance as to whether the hours would average more than those of most workers in like employment or be so much more irregular as to be substantially less favorable.
As indicated in the discussion of wages, the hours of work in some occupations are also subject to seasonal fluctuations. In the needle trades, for example, the workers generally put in long hours during the rush season, particularly in the fall. During dull periods when work is slow, many are laid off and others work only a short week; that is, less than the normal weekly schedule. In such cases, it is generally best to compare the hours of the work offered with those prevailing on the basis of the normal work schedule and to make allowance for temporary or seasonal fluctuations. Again, the experience of other workers in the establishment may offer some guidance as to the extent of the fluctuations in the job offered as compared with those ordinarily to be expected and whether the hours would on the whole be no longer than those of most workers in similar employment.
Some care may have to be exercised to distinguish between temporary changes and fluctuations of this kind and permanent increases or reductions in the hours of work. The distinction would be especially important if the wage determination is made only in terms of wage rates since an offer of work which regularly involves shorter hours than those prevailing would ordinarily result in lower earnings even if the rates were the same.
In addition, any general change in the regular hours of a substantial number of workers in the occupation may also affect the prevailing hours determination. Thus, if the hours of a considerable number of workers are increased, reexamination may reveal, for example, that a greater number are now employed on a 44-hour schedule than any other whereas a 44-hour week had previously prevailed. Similarly, if the hours of most of the workers in the occupation are reduced an offer of work at the hours which previously prevailed may not be substantially less favorable than those currently prevailing.
The hours of the work offered may also be substantially less favorable if they are less convenient than those prevailing in the locality for similar work. Thus, if most workers in the occupation work a 40-hour week on the basis of 5 8-hour days with Saturday and Sunday off, an arrangement whereby the worker would be required to put in 5 7-hour days and 5 hours on Saturday may be substantially less favorable to the individual than that prevailing because it leaves him only 1 day a week free even though the total number of hours is no longer than those of most workers.
Similarly, second or third shift work would generally be substantially less favorable if most of the workers in the occupation were employed on the first shift. It is because the second and third shifts are recognized as less convenient by both employers and employees that differentials are frequently paid for such work. Special payments of this kind, like extra pay for evening or holiday work, do not generally affect the hours determination. However, where the shift differential takes the form of shorter hours for equal pay, longer hours than those prevailing for second or third shift work might well be held substantially less favorable to the claimant.
There would, of course, be no question under the prevailing conditions of work provisions as to whether any shift was substantially less favorable than another if a relatively equal number of workers were employed on all shifts. In such circumstances no one shift could be said to prevail. If, however, a fairly equal number are employed on the first and second shift, an offer of work on the third shift might well be deemed substantially less favorable to the worker than the prevailing hours of work--unless such workers are generally hired on the least desirable shift and earn the right to move up to an earlier shift only as they acquire seniority. In the latter instance, the fact that the right to work on an earlier shift depends on the worker's seniority would itself be a condition of work. In such cases, determination of the prevailing arrangement of hours would be a matter of determining the shift on which the workers in the occupation are customarily hired in the locality rather than the shift on which the greater number are currently employed.
Subject to the same exception, a split shift which involves working at two different times of the day, or a swing shift which involves changing over between two different shifts at stipulated weekly intervals, would generally be substantially less favorable to the worker than the prevailing arrangement of hours if a straight shift prevailed; and a rotating three-shift arrangement would generally be substantially less favorable if either a straight shift or a swing shift prevailed. Other factors such as the hours involved and the claimant's circumstances may also enter into the determination, however. Thus, if the workers in the occupation are generally hired on the third shift, a rotating shift involving change-over between the third, second, and first shifts might not be substantially less favorable to the individual provided he was able to work on all three shifts and the constant change in hours would not affect him adversely.
Whether lesser differences such as the time a shift begins and ends or in the length of the lunch hour, etc., render the hours of work substantially less favorable to the individual also depends on the nature and extent of the difference and on the claimant's circumstances. Thus, if the claimant would be required to report to work at 6:30 a. m. whereas most workers in like employment did not begin to work until 9:00 a. m., the hours might well be held substantially less favorable than those prevailing. But a difference of a half hour or three-quarters of an hour in the time the shift started might not be material if it would adversely affect the claimant. In other cases the omission of rest periods granted most workers in like employment and differences in the length of the lunch hour or the starting hour may be compensated by other circumstances such as the fact that the workers are seated on the job or the existence of lunchroom facilities on the premises.
Generally, though, it will not be necessary to go into questions of this kind. The hours characteristic of the occupation in the particular locality will usually govern the decision as to whether an inconvenient shift or arrangement of hours is substantially less favorable to the individual.
As ordinarily used, the phrase "conditions of work" refers to the provisions of the employment agreement, both express and implied, and the physical conditions under which the work is done pursuant to the agreement. It is also applied at times to conditions which arise from actual work on the job as a result of laws and regulations which are not within the employer's control. So interpreted, the phrase "conditions of work" includes such factors as coverage by the State workman's compensation and unemployment compensation acts and the Federal old-age and survivors insurance provisions.
Under either interpretation, the phrase encompasses not only wages and hours but such other factors as:
Moreover, while the list set forth above by way of illustration of the more common factors which may be important in various occupations and localities is extensive, it is by no means all inclusive. There are many other factors which may be important in certain occupations and localities.
Thus in outdoor employments, if it appears that the claimant would be required to work in all kinds of weather, it may be important to ascertain if most workers in like employment in the locality are required to be on the job regardless of the weather and if some shelter or protection is generally provided. In inspection jobs and in the case of stock chasers and many other employments, the weight of the parts or materials the worker may have to lift without mechanical aid may be important. In longshoreman's work and in the case of deliverymen and movers the size of the crew is often a matter of negotiation.
In the needle trades, questions may arise as to the state of repair in which machines are kept or whether the worker would be required to fix his own machine, since a poorly adjusted machine results in spoilage and lower earnings at piece rates and the time spent repairing the machine is lost to the worker. In the textile industry, the number of machines or bobbins the worker is required to tend is frequently an issue. In coal mining the height of the chamber in which the work is done, the presence of water or gas, the frequency with which the mine is inspected, and the amount of timbering or other nonproductive work required may be important.
Because of the innumerable variations in the conditions under which workers are employed in various occupations and localities, and because many of the conditions other than wages and hours are so closely interrelated with the nature of the work, it is not possible to discuss them without going into the details of particular trades and industries. Nor can any generalization be made about the relative importance of many of these conditions without considering them in relation to each other. Thus the lack of a guaranteed minimum weekly wage may be a technical rather than a material difference if the worker would in all probability regularly earn as much or more than the amount guaranteed to most workers in like employment in the locality. Similarly, the importance of a seniority provision would depend on whether it only dictated the order in which workers in the occupation would be laid off or also determined promotions and transfers from one department or shift to another.
In general, however, the question under the prevailing conditions of work provisions as to conditions other than wages or hours is whether the conditions of the work offered are substantially less favorable to the claimant than those prevailing in any important respect. The claimant is not subject to denial of benefits for refusal of work if the wages, hours, or any other material condition or combination of conditions of the work offered is substantially less favorable to him than those prevailing in the locality for similar work.
If there is reason to believe that the conditions of the work offered are less favorable than those prevailing for similar work in the locality in any important respect, it is for the agency to investigate. The issue in each case must be decided on the basis of all the relevant facts and the best information available.