General Administrative Letter 1-00
DIRECTIVE: GENERAL ADMINISTRATION LETTER NO. 1-00
TO: ALL STATE EMPLOYMENT SECURITY AGENCIES
FROM: LENITA JACOBS-SIMMONS Deputy Assistant Secretary
SUBJECT: Availability and Use of Occupational Employment Statistics Survey Data for Alien Labor Certification Prevailing Wage Purposes
1. Purpose. To make State staff aware of the availability of new Occupational Employment Statistics Survey (OES) data and to provide information on how it is to be used in light of the changes to prevailing wage determination practices as promulgated by the American Competitiveness and Workforce Improvement Act (ACWIA); and to transmit Prevailing Wage Policy (Attachment A) which provide policy and procedural guidance clarifying the intent of General Administration Letter (GAL) No. 2-98, particularly with regard to the use of employer-provided wage data.
2. References. 20 CFR part 655, subpart A; 20 CFR part 655, subparts H and I; 20 CFR part 656; Technical Assistance Guide (TAG) No. 656, Labor Certifications; GAL No. 2-98, Prevailing Wage Policy for Nonagricultural Immigration Programs (October 31, 1997); and GAL No. 2-99, Availability and Use of Occupational Employment Statistics Survey Data for Alien Labor Certification Prevailing Wage Purposes (April 23, 1999).
3. Background. The Bureau of Labor Statistics= (BLS) regional offices have recently provided new data files to Employment Security Agencies which contain the updated OES data. States are to use this data in making prevailing wage determinations during Calendar Year 2000. As was the case in 1999, there are two data sets. One set is composed of the ALC__98f.dbf and *.dat files, those with the prefix AALC and referred to as the AALC database. These files will be used to determine prevailing wages for occupations in most industries. The second set is composed of the EDC__98f.dbf and *.dat files, those with the prefix AEDC and referred to as the AEDC database. These files are to be used to determine prevailing wages in industries identified in ACWIA, as discussed below. These data sets should be used for making prevailing wage determinations immediately upon receipt.
This years data are the third year of OES estimates and have been developed using 1996, 1997, and 1998 survey data that, when combined, cover approximately 1,200,000 sample units. As expected, wages for use in 2000 differ from those used in 1999. One of the results of the new sample is that the number of occupation/location cells whose data is based on data only obtained within the specific location has again increased from last year.
The ACWIA, enacted as part of the omnibus budget for Fiscal Year 1999, increased the H-1B quota and made several other changes to the program, including a change for the calculation of prevailing wages in all labor certification programs for all occupations with specific types of employers. The ACWIA states: Section 212 (8 U.S.C.1182) is amended by adding at the end the following:
The new law went beyond the recently published regulations referring to researchers in colleges and universities and selected research facilities. The ACWIA indicates that all occupations in institutions of higher education, a related or affiliated nonprofit entity, nonprofit research organization, or governmental research agency, should be provided prevailing wages using the AEDC database or use the prevailing wage based on a sample of similar institutions.
The OES survey collected wage data on occupations found in institutions of higher education (SIC 822). This is the AEDC database (files listed as EDC__98xxf.dbf and *.dat where xx is the appropriate State postal code). The AEDC database is to be used to determine the prevailing wage for occupations in organizations identified in ACWIA above (institutions of higher education, a related or affiliated nonprofit entity, nonprofit research organization, or governmental research agency).
The Davis-Bacon and Service Contract Act wages are not to be used as the prevailing wage for occupations in organizations identified in ACWIA because of the specific wording in the Act regarding the way in which prevailing wages are to be determined for these organizations. However, if a negotiated union agreement exists for the requested job in the facility where the job is being requested, the negotiated wage is to be used as the prevailing wage.
We are currently discussing with BLS the possibility of obtaining wage data for Governmental research organizations. Although BLS has data from OPM on Federal wages, we must determine if we can extract from that data those wages paid in organizations where the primary function is research. Until that information can be provided, the EDC database should be used to provide prevailing wage determinations in these organizations as well.
For users of ALCATS, that computer program has been revised to allow the user to select the use of the ALC or EDC database, as appropriate. The application is available to download by accessing the internet site http://www.flcdatacenter.com/ and following the appropriate links to the ALCATS area. State Agency staff that would like to obtain a CD containing the application may obtain it either by using the order form available at that site, or by calling Gary Ray at 801-526-9455.
5. Action Required. State Administrators are requested to:
6. Inquiries. Inquiries regarding this GAL should be addressed to your Regional Office.
1. Must an employer use the OES wage provided by the SESA?
No. An employer can use a private survey provided the survey meets the seven criteria outlined in item J of GAL 2-98.
2. What are the seven criteria?
(i) The data must have been collected within 24 months.
(ii) If a published survey, the survey must have been published within 24 months.
(iii) The survey must reflect the area of intended employment.
(iv) The employer job description must adequately match the survey job description.
(v) The survey must be across industries that employ workers in the occupation.
(vi) The wage determination must be based on an arithmetic mean.
(vii) The survey must identify a statistically valid methodology that was used to collect the data.
Because there are many ways to categorize and calculate salaries, the Department made available the option of using a private survey. Therefore, if an attorney advises an employer, or if an employer elects, to use a private survey, SESAs and regional offices should accept it and review it according to the seven criteria. If the alternative survey meets the seven criteria, it should be accepted.
4. If a SESA accepts a private survey, is that survey determination then the new wage for that occupation in the given area?
No. A private survey is good one time for the employer that submits the survey. Other employers will be issued the OES wage unless they submit a survey for review.
5. If a private survey is rejected by a SESA, would the employer be entitled to submit a new survey?
Yes. However, this should not be confused with a challenge to a prevailing wage determination made by a SESA. If a private survey is rejected by a SESA and the employer challenges the SESA=s rejection, whether through the ES complaint system under the H-1B program, or through the regional certifying officer under the permanent program, the challenge is limited to the validity of the survey rejected by the SESA. The prevailing wage challenge process is not intended to serve as a forum for an employer to submit multiple private surveys.
6. Can an employer submit more than one survey at a time?
No. Only one survey will be evaluated at a time for a given employer.
7. Must the private survey mirror the OES survey?
No. It is not necessary that the private survey mirror the OES survey. It is not necessary to use the identical geographic region, cross reference, etc.
B. Sample and Sample Size
8. What is the minimum sample size that will be accepted?
There is no minimum sample size. It will depend on the occupation, the area of intended employment, and the area surveyed. The key determinant is whether the sample is representative. The private survey must include a representative sample of workers in the occupation in the area of intended employment.
9. Some private surveys do not differentiate types of positions within a given occupation. (Example: engineers can include electronic, mechanical, industrial). Must the private survey distinguish within a given occupation?
No. The private survey does not have to be limited to the specific occupational classification within a field. A survey that includes all engineers, in an effort to capture a broader sample, may be accepted provided the survey does not also provide usable wage rates for the specific occupational classification. For example, if the job opportunity for which a prevailing wage determination is sought is a nuclear engineer, a survey for engineers would be acceptable provided that the survey does not include usable wage data for the specific occupational classification of nuclear engineers. There is no requirement that a private survey include a cross-sample of occupational types. There simply must be an adequate match of job duties, not a precise and exact match.
10. Does a survey have to utilize a random sampling procedure or is it adequate to have a representative sampling of employees obtained through other statistical methods? A random sampling is not an absolute requirement but the survey has to reflect a representative sampling of similarly employed workers in the area surveyed.
C. Area of Intended Employment
11. The survey must reflect area of intended employment. Must the employer survey use the same geographic area as the government survey?
Not necessarily. In all cases, if the employers survey is limited to the area of intended employment, it meets the geographic requirement of GAL 2-98. If the survey is broader than the area of intended employment, the employer must establish that there were not sufficient workers in the area of intended employment, thus necessitating the expansion of the area surveyed. The area of intended employment should not be expanded beyond that which is necessary to produce a representative sample.
12. Is it permissible for a private survey to encompass an area different from or larger than an Metropolitan Statistical Area (MSA), a Primary Metropolitan Statistical Area (PMSA) or a Balance Of State (BOS) if all sources are obtained from locations within normal commuting distance?
Yes. Under the Departments regulations, normal commuting distance is always the first order of inquiry in determining the scope of the area of intended employment.
13. Will a survey that includes employers from another state be acceptable if the out-of-state employers surveyed are located within normal commuting distance of the intended place of employment?
14. Will a survey utilizing a Consolidated Metropolitan Statistical Area (CMSA) be accepted if all points on the survey are within normal commuting distance of the employer?
15. Must an employer survey be limited to contiguous area if the OES survey utilizes a level 2 geographical area (contiguous areas)?
No, as long as the employer presents documentation to establish that it could not obtain an adequate sample from within the level 2 area.
16. May a CMSA be a reasonable alternative if the OES survey utilizes a level 2 geographic area (contiguous areas)?
17. May a private survey use a larger geographic area than the OES survey?
Possibly. A private survey that uses a larger geographic area might be acceptable. If the private survey uses a geographic area beyond the MSA, PMSA or BOS, than the employer must explain why it was not possible to obtain an adequate sample size within the MSA, PMSA or BOS. For example, a survey company might not be able to obtain an adequate sample within the MSA and therefore would not be able to provide a wage determination for the specific occupation at a level equal to or smaller than the MSA. In this case, the employer should document why it was not possible to obtain a representative sample within the MSA based upon standard survey practices. This is an example of a situation in which an attempt was clearly made to obtain data within the MSA, and as a result the survey should not be rejected on the basis that it encompasses a larger geographic area than the OES survey.
18. Is a survey acceptable if the employer uses a smaller geographic area than the OES survey?
As stated above, a private survey need not mirror the OES survey. If the employer was able to obtain an adequate and representative sample within a smaller geographic region than that used for the OES survey, and that area is representative of the area of intended employment, then the private survey should be accepted.
19. Is a private survey acceptable if it crosses state lines?
Yes. If the survey was done within normal commuting distance, or if crossing state lines was necessary to obtain an adequate sample.
D. Cross Industries
20. The survey must be across industries. Is there a minimal number of industrial codes that must be included in the cross-industry survey?
No. It simply needs to be a representative sampling of workers in the occupation in cross industries.
21. May an employer use an industry-specific survey if an occupation is only present within several industries?
Generally speaking an employer provided survey must be across industries. However, if an occupation is found predominantly in one industry, an industry-specific survey would be acceptable provided that it is based on a representative sampling of workers in that occupation.
E. Skill Levels
22. When matching job descriptions, can a private survey include more than two skill levels and still be acceptable?
23. Must the employer survey reflect wages for Level I and Level II occupations as defined in GAL 2-98?
No. It is not necessary that a private survey distinguish occupational wage levels as is done under the OES survey. The level selected by the employer from their survey must be the one that most closely matches the level of experience or responsibility in the job offer. A private survey that provides one overall average for an occupation is acceptable provided that the survey does not also include usable wage data for varying levels of skill or responsibility within the occupation. In such cases, the arithmetic mean for the skill level that most closely matches the employers job opportunity should be used.
F. Time Frames
24. The employer must submit the most recent survey and it must not be more than 24 months old. Is the 24 months to be calculated from the time the survey is submitted to the SESA or region, or from the time a determination is made as to the survey acceptability?
Generally, the survey must be the most recent available and must not be more than 24 months old at the time it is submitted by the employer. If a survey is published more frequently than annually, the employer may use any edition of the survey with usable wage data that was published within the year prior to the date it is submitted.
25. What is the time frame within which DOL must respond (accept or reject) to the employers request that a private survey be accepted?
A response should be issued in writing in a timely manner, preferably within 30 working days of receipt of the survey.
Attachment B to GAL No. 1-00 HIGHER EDUCATION ACT OF 1965
Sec. 101(a) As used in this Act --
(a) INSTITUTION OF HIGHER EDUCATION- For purposes of this Act, other than title IV, the term
Last updated: 06-04