Fair Labor Standards Act: White-Collar Exemptions in the Modern Work
Place (Letter Report, 09/30/1999, GAO/HEHS-99-164).

Pursuant to a congressional request, GAO provided information on
employer compliance with the white-collar exemptions under the Fair
Labor Standards Act (FLSA), focusing on: (1) how many employees are
covered by the white-collar exemptions and how the demographic
characteristics of these employees have changed in recent years; (2) how
the statutory and regulatory requirements have changed since the
enactment of FLSA; (3) the major concerns of employers regarding the
white-collar exemptions; (4) the major concerns of employees regarding
the white-collar exemptions; and (5) the possible solutions to the
issues of concern raised by employers and employees.

GAO noted that: (1) in 1998, between 20 and 27 percent of the full-time
U.S. workforce--or 19 to 26 million workers--were executive,
administrative, or professional employees covered by white-collar
exemptions of the FLSA; (2) in recent years the percentage of employees
covered by these exemptions has been increasing; (3) the number of
employees working in certain service industries nearly doubled between
1983 and 1998, and there is a higher percentage of white-collar
employees in the service sector than in other sectors of the economy;
(4) overall, the workforce covered by the exemptions also became
increasingly female--the proportion of women increased from 33 percent
in 1983 to 42 percent in 1998; (5) in the 16 years following the 1938
enactment of the FLSA, the Department of Labor (DOL) established the key
regulatory tests defining whether an employee can be classified as an
exempt white-collar worker; (6) these tests included the salary basis
test--the requirement that exempt white-collar workers be paid a salary,
not an hourly wage--as well as the various salary-level and duties
tests; (7) since 1954, major statutory and regulatory changes to the
white-collar exemptions have been few, and primarily limited to
increases in the salary-test levels and to changes to coverage of
specific types of employees; (8) employers worried about potential
liability for violations of the salary-basis test; (9) employers also
believed that the regulations limiting the exemptions to white-collar
nonproduction employees did not take into account the effect of modern
technology on employment; (10) employers complained that the parts of
regulatory duties tests that call for independent judgment and
discretion on the part of administrators and professionals led to
confusing and inconsistent results in classifications of similarly
situated employees; (11) employee representatives believed that
inflation has severely eroded the salary-level limitations originally
envisioned by the DOL regulations; (12) the representatives contended
that the duties test for executive employees has been oversimplified,
leading to inadequate protection of low-income supervisory employees;
(13) although various proposals have been advanced to address the
concerns raised in this report, the conflicting interests of employers
and employees have made resolution difficult; and (14) to resolve these
issues, the desire of employers for clear and unambiguous regulatory
standards must be balanced with that of employees for fair and equitable
treatment in the work place.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  HEHS-99-164
     TITLE:  Fair Labor Standards Act: White-Collar Exemptions in the
	     Modern Work Place
      DATE:  09/30/1999
   SUBJECT:  Labor law
	     Industrial relations
	     Wage surveys
	     Overtime compensation
	     Labor statistics
	     Personnel classification
IDENTIFIER:  Census Bureau Current Population Survey

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************

Cover
================================================================ COVER

Report to the Subcommittee on Workforce Protections, Committee on
Education and the Workforce, U.S.  House of Representatives

September 1999

FAIR LABOR STANDARDS ACT -
WHITE-COLLAR EXEMPTIONS IN THE
MODERN WORK PLACE

GAO/HEHS-99-164

FSLA and White-Collar Exemptions

(205387)

Abbreviations
=============================================================== ABBREV

  BLS - Bureau of Labor Statistics
  CPS - Current Population Survey
  DOL - Department of Labor
  FLSA - Fair Labor Standards Act

Letter
=============================================================== LETTER

B-283016

September 30, 1999

The Honorable Cass Ballenger
Chairman
The Honorable Major R.  Owens
Ranking Minority Member
Subcommittee on Workforce Protections
Committee on Education and the Workforce
House of Representatives

The Honorable Bill Goodling
Chairman
The Honorable William Clay
Ranking Minority Member
Committee on Education and the Workforce
House of Representatives

After more than 60 years, the Fair Labor Standards Act (FLSA) remains
the primary federal statute setting the minimum wage and hour
standards applicable to most American workers.  Since its enactment
in 1938, the industrial profile of the American economy has shifted
dramatically, changing from predominantly manufacturing to
increasingly service-oriented.  Critics of the FLSA claim that this
shift, as well as the increased use of sophisticated technology, have
left the FLSA and its regulations outdated and in need of revision. 

One area of concern involves the so-called "white-collar" exemptions
of the FLSA.  The Act limits the normal work-week to 40 hours,
requiring most employers to pay hourly overtime wages to employees
who work longer than 40 hours.  However, under section 13(a)(1) of
the Act, employees working in a "bona fide executive, administrative,
or professional capacity" are exempted from the wage and hour
standards.  These white-collar employees need not be paid overtime
premium pay for a work-week longer than 40 hours. 

Employers from both the private sector and state and local
governments have focused their criticisms on Department of Labor
(DOL) regulations that define the "exempt" white-collar employees. 
Under the FLSA, DOL is responsible for setting the criteria for these
exemptions, and historically it has formulated specific regulatory
tests based on the accumulated experience of employers, employees,
and its own field staff with work-place issues.  Currently, employees
must meet each of three tests to be classified as exempt white-collar
workers:  (1) the employee must be paid a salary, not an hourly wage
(the salary-basis test); (2) the amount of the employee's salary must
indicate managerial or professional status (the salary-level tests);
and (3) the employee's job duties and responsibilities must involve
managerial or professional skills (the duties tests). 

In response to your request for information on employer compliance
with the white-collar exemptions under the FLSA, this report focuses
on five questions:  (1) How many employees are covered by the
white-collar exemptions and how have the demographic characteristics
of these employees changed in recent years?  (2) How have the
statutory and regulatory requirements changed since the enactment of
the FLSA?  (3) What are the major concerns of employers regarding the
white-collar exemptions?  (4) What are the major concerns of
employees regarding the white-collar exemptions?  (5) What are
possible solutions to the issues of concern raised by employers and
employees?  We performed our work in accordance with generally
accepted governmental auditing standards from November 1998 through
June 1999.  Our scope and methodology are presented in appendix I. 

   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

In 1998, between 20 and 27 percent of the full-time U.  S. 
workforce--or 19 to 26 million workers--were executive,
administrative, or professional employees covered by white-collar
exemptions of the FLSA.\1 In recent years the percentage of employees
covered by these exemptions has been increasing.  The number of
employees working in certain service industries\2 nearly doubled
between 1983 and 1998, and there is a higher percentage of
white-collar employees in the service sector than in other sectors of
the economy, such as manufacturing.  Overall, the workforce covered
by the exemptions also became increasingly female--the proportion of
women increased from 33 percent in 1983 to 42 percent in 1998.  In
addition, in 1998, workers subject to the white-collar exemptions
were more than twice as likely as nonexempt workers to work
overtime--44 percent of exempt employees worked more than 40 hours in
a work-week, and about one-third of those worked more than 50 hours
in a work-week. 

In the 16 years following the 1938 enactment of the FLSA, DOL
established the key regulatory tests defining whether an employee can
be classified as an exempt white-collar worker.  These tests included
the salary-basis test--the requirement that exempt white-collar
workers be paid a salary, not an hourly wage--as well as the various
salary-level and duties tests.  Since 1954, major statutory and
regulatory changes to the white-collar exemptions have been few, and
primarily limited to increases in the salary-test levels and to
changes to coverage of specific types of employees.  In recent years,
for example, the salary-basis test has been adjusted for state and
local government employees, and higher-wage computer programmers were
included in the exemption. 

In general, employers we contacted were concerned that the regulatory
tests were too complicated and outdated.  Specifically, their
concerns included the following: 

  -- Employers worried about potential liability for violations of
     the salary-basis test.  While DOL viewed the test as being a
     highly accurate indicator of managerial and professional status,
     in recent years it has been the focus of legal suits brought
     collectively by groups of managerial and professional employees
     against their employers.  Our review of federal cases and
     discussions with employers showed continuing uncertainties and
     difficulties with the test. 

  -- Employers also believed that the regulations limiting the
     exemptions to white-collar nonproduction employees did not take
     into account the effect of modern technology on employment.  For
     example, they pointed to highly skilled and well-paid
     technicians who did not qualify as exempt professionals, but who
     performed essentially the same job as exempt engineers with the
     required academic degrees. 

  -- Finally, employers complained that the parts of regulatory
     duties tests that call for independent judgment and discretion
     on the part of those classified as administrators and
     professionals led to confusing and inconsistent results in
     classifications of similarly situated employees.  Our
     discussions with DOL investigators and review of compliance
     cases indicated that this part of the duties test involved
     difficult and sometimes subjective determinations, and that it
     was a source of contention in DOL audits. 

Employee representatives, on the other hand, were most concerned
about preserving work-hour limitations for employees, and believed
that the regulatory tests, as applied today, were not sufficient to
adequately restrict the use of the exemptions by employers. 
Specifically, they cited the following concerns: 

  -- Employee representatives believed that inflation has severely
     eroded the salary-level limitations originally envisioned by the
     DOL regulations.  The regulations create three levels of
     regulatory duties tests, depending on employees' salaries. 
     Under the regulations, the lower the employee's salary, the
     greater the limitations on the use of the exemptions.  However,
     the regulations do not provide for automatic or periodic
     adjustments of the salary levels, and the levels have not been
     changed since 1975.  To fully account for inflation between 1975
     and 1988, the salary levels would have to be increased about
     threefold.  As a result of the increase in salaries over that
     period, almost all full-time employees in 1998 were covered by
     the least-restrictive regulatory duties test--leaving more
     people than ever who potentially fall under the white-collar
     exemptions. 

  -- The representatives contended that the duties test for executive
     employees has been oversimplified, leading to inadequate
     protection of low-income supervisory employees.  Our review of
     federal case law and DOL compliance cases indicated that it is,
     in fact, difficult to challenge exempt classifications if
     employees supervise two or more full-time employees and spend
     some time--even if minimal--on management tasks. 

Although various proposals have been advanced to address the concerns
raised in this report, the conflicting interests of employers and
employees have made resolution difficult.  Some proposals would, for
example, eliminate the salary-basis test or raise the salary-test
levels.  However, for every proposal--even those with consensus, such
as increasing the salary-test levels--there are competing interests
to be considered.  To resolve these issues, the desire of employers
for clear and unambiguous regulatory standards must be balanced with
that of employees for fair and equitable treatment in the work place. 
Although DOL established the regulatory tests by balancing these
competing interests, these same interests have made DOL reluctant to
alter the current regulatory structure.  In the last 45 years, DOL
has adjusted the FLSA regulations only in a piecemeal fashion to meet
the needs of particular types of employers and employees.  Given the
economic and work place changes over this period, a more
comprehensive look at these regulations is necessary to determine
whether a consensus could be achieved on how to amend the regulations
to better suit the modern work place.  This report recommends that
the Secretary of Labor comprehensively review current regulations and
restructure white-collar exemptions to better accommodate today's
work place and to anticipate future work place trends. 

--------------------
\1 Our estimate includes only those employees who would most likely
be properly classified as exempt workers under the DOL regulations. 
It may not include all employees who are classified as exempt workers
by their employers. 

\2 These industries included four types of service occupations from
the Current Population Survey (CPS) industry codes:  business and
repair, personal, entertainment and recreation, and professional and
related services.  For definitions of other industries discussed in
this report, see app.  I. 

   BACKGROUND
------------------------------------------------------------ Letter :2

The FLSA sets the minimum wage most employers must pay their
employees and the maximum hours--40 per week--most employees can work
without receiving extra, overtime premium pay (at time-and-one-half
the regular rate).  In addition, the FLSA specifies which workers are
exempt from these requirements.  Although numerous categories of
workers are exempt from these requirements,\3 the largest group of
exempt workers includes employees classified as executives,
administrators, or professionals under section 13(a)(1) of the Act. 
These are sometimes called the white-collar exemptions, although not
all white-collar employees are exempt. 

The FLSA was enacted to address problems associated with substandard
working conditions by establishing a floor on wages and a ceiling on
hours, beyond which the employer was required to pay extra wages. 
The purpose of the overtime provision was to shorten the work-week to
a more reasonable 40 hours.  This was expected to result in less
employee fatigue, fewer accidents, higher productivity and
efficiency, and more employee time for education and family duties. 
By requiring overtime premium pay, it was expected that employers
would hire more workers to avoid the extra wage costs, and that
workers would be assured additional pay to compensate them for the
burden of a work-week in excess of 40 hours.  The Minimum Wage Study
Commission of 1981\4 justified the exemption of executives,
administrators, and professionals from the protections of the FLSA in
part because these employees were associated with higher base pay,
higher promotion potential, and greater job security, making them
different from other employees.  Moreover, the nature of their
jobs--managerial and professional--precluded the potential for the
job expansion desired in other types of employment (that is, hiring
more workers to perform the additional hours of work). 

For employers and employees, the practical consequences of the exempt
worker classification can be very important.  An exempt employee may
be required to work as many hours as it takes to complete a task. 
Although this may be more than 40 hours per week, the employee will
not be entitled to overtime premium pay for the hours exceeding 40. 
Thus, an exempt financial manager may be required to work 60 hours a
week and be paid a set weekly salary.  On the other hand, a nonexempt
bookkeeper may also be required to work 60 hours per week, but must
be paid at a premium hourly wage for 20 hours (the number exceeding
40 per week) in addition to a set weekly salary.\5

Ever since the FLSA was enacted, the interests of employers in
expanding the white-collar exemptions as broadly as possible have
competed with those of employees in limiting use of the exemptions. 
In 1940, for example, DOL reported that groups representing employers
argued for broader use of the exemptions to allow management
training, to increase flexibility in work-hour scheduling, and to
ensure a stable weekly pay for employees.  At the same time, employee
representatives argued against broader use of the exemptions, trying
to reduce the potential for abuse and exploitation of workers. 

Balancing the competing interests of employers and employees, DOL
established specific regulatory tests that must be met before an
employee can be classified as an exempt white-collar\6 worker.  In
general, there are three major parts to these tests: 

  -- First, the employee must be paid on a salary basis, not at an
     hourly rate.  This means that the employee must be paid a
     guaranteed amount each pay period, independent of the number of
     hours that the employee has actually worked and the quality and
     quantity of work performed. 

  -- Second, the employee must be paid at least a specified base
     salary level that indicates managerial or professional status. 
     DOL regulations include different salary levels.  One is a base
     level for each type of exempt white-collar worker--executive,
     administrative, or professional--below which workers are assumed
     to be nonexempt and covered by the FLSA minimum wage and
     overtime requirements.  The other is a higher salary level,
     above which employees will likely be exempt if their primary
     duties are managerial or professional. 

  -- Third, the employee must have duties and responsibilities
     associated with managerial or professional work.  Generally,
     such duties must include appropriate independent judgment and
     discretion.  However, depending upon the employee's salary
     level--whether it is above or below the highest salary
     level--DOL regulations call for either closer scrutiny (with a
     long, detailed test) or not as much scrutiny (with a short,
     limited test) of the nature of the employee's duties. 

The regulatory tests vary among the three categories of
employees--executive, administrative, and professional.  Table 1
summarizes the major tests required for each type of exemption.  For
each category of employee, we identify the salary levels included in
the regulations and the associated duties test.  We refer to the
lower salary as the base salary, and the applicable duties test as
the long test.  The higher salary level is referred to as the upset
test, and the applicable duties test as the short test. 

                                          Table 1
                          
                          Summary of Current Regulatory Tests for
                               Executive, Administrative, and
                                Professional FLSA Exemptions

                                                           Upset
                                                           salary
                           Base salary                     (triggers
                           (triggers                       short
               Paid a      long duties                     duties
Employee type  salary      test)         Long duties test  test)       Short duties test
-------------  ----------  ------------  ----------------  ----------  -------------------
Executive      Yes         $155 per      Various           $250 per    (1) Must supervise
                           week          indicators,       week        two or more
                                         including a                   employees, and (2)
                                         primary duties                primary duty must
                                         test and a                    be management
                                         requirement that
                                         no more than 20
                                         percent of work
                                         (or 40 percent
                                         if in retail or
                                         service) involve
                                         nonmanagerial
                                         work

Administrativ  Yes; also   $155 per      Primary duties    $250 per    (1) Primary duty
e              may be      week          test including    week        must involve office
               paid on a                 the percentage                or nonmanual (or
               fee basis                 limitations on                staff) work
                                         nonexempt work,               directly related to
                                         plus other                    management, and (2)
                                         indicators of                 work must require
                                         administrative                discretion and
                                         responsibilities              independent
                                                                       judgment

Professional   Yes; also   $170 per      Primary duties    $250 per    Either (1) must
               may be      week          test including    week        have requisite
               paid on a                 the percentage                academic degree and
               fee basis                 limitations on                job must require
                                         nonexempt work,               consistent exercise
                                         plus other                    of discretion and
                                         indicators of                 independent
                                         professional                  judgment, or (2)
                                         responsibilities              must involve
                                                                       original and
                                                                       creative work
                                                                       requiring
                                                                       invention,
                                                                       imagination, or
                                                                       talent in
                                                                       recognized field
------------------------------------------------------------------------------------------
Source:  GAO analysis of DOL regulations. 

--------------------
\3 Currently, section 13(a) lists 10 other categories of workers (in
addition to managers and professionals) as exempt from both the
minimum wage and maximum hours provisions of the FLSA.  These include
diverse groups of employees, such as babysitters and those working at
recreational establishments. 

\4 The legislative history for the FLSA contains no explanation for
the exemption. 

\5 Salaried workers may be either exempt or nonexempt; being paid a
salary is not determinative of exempt status. 

\6 DOL does not refer to a white-collar exemption; the exemption is
referred to routinely as covering executive, administrative, and
professional employees. 

   NUMBER OF WHITE-COLLAR
   EXEMPTIONS INCREASES WITH
   GROWTH OF SERVICE SECTOR
------------------------------------------------------------ Letter :3

We estimate that between 19 and 26 million full-time wage and salary
workers\7

were covered by the white-collar exemptions in 1998.\8 \9 This
amounts to 20 to 27 percent of the full-time labor force.  Based on
the high estimate of 26 million, our estimate represents an increase
of 9 million workers over our 1983 estimate of 17 million exempt
full-time wage and salary workers (see table 2).  For a detailed
description of the methods used to obtain these estimates, see
appendix I. 

                                         Table 2
                         
                           Estimates of Full-Time White-Collar
                             Workers Exempt in 1983 and 1998

                                           Covered by white-collar exemptions
                               ----------------------------------------------------------
                                      High estimate                  Low estimate
                               ----------------------------  ----------------------------
                  Total full-                 Percentage of                 Percentage of
                time wage and     Number of  full-time wage     Number of  full-time wage
               salary workers     employees      and salary     employees      and salary
Year             (millions)\a    (millions)         workers    (millions)         workers
-------------  --------------  ------------  --------------  ------------  --------------

 1983                      71            17             24%            12             17%

 1998                      96            26             27%            19             20%
-----------------------------------------------------------------------------------------
Notes:  Includes employees exempt under sec.  13(a)(1) of the FLSA. 
29 C.F.R.  541 defines those employees classified as executive,
administrative, professional, or outside sales workers.  Outside
sales workers are not included in this analysis.  Please see app.  I
for a discussion of these estimates. 

\a Wage and salary employment numbers are from the CPS Outgoing
Rotations Data analysis and match the Bureau of Labor Statistics
(BLS)-published Employment and Earnings annual averages. 

Source:  CPS Outgoing Rotations Data for 1983 and 1998. 

Much of the growth in exempt workers can be attributed to the growth
in the service sector of the economy.  In 1998, the service
industries employed 24 million full-time workers--nearly doubling
from 13 million workers in 1983.  All sectors of the labor market saw
some growth in the number of workers between 1983 and 1998;\10
however, no other sector has grown as rapidly in the last 15 years. 
As figure 1 shows, in 1998 one-quarter of all full-time workers held
jobs in the service sector, which makes it the largest employment
sector. 

   Figure 1:  Percentage of
   Full-Time Wage and Salary
   Workers in 1983 and 1998 by
   Industry

   (See figure in printed
   edition.)

Notes:  The sampling errors for the estimates in this figure do not
exceed plus or minus 0.5 percentage points at the 95% significance
level.

Service industries included four types of service occupations from
the Current Population Survey (CPS) industry codes:  business and
repair, personal, entertainment and recreation, and professional and
related services.  For definitions of other industries discussed in
this report, see app.  I. 

Source:  CPS Outgoing Rotations Data for 1983 and 1998. 

In addition to growing rapidly, the service sector also has a higher
proportion of exempt workers than other sectors and is responsible
for much of the growth in the exempt population.  Not only has the
service sector grown by 11 million full-time workers in the last 15
years, but the number of exempt workers in the service sector has
increased by 3.6 million.\11 Over the last 15 years, the increase in
the number of full-time workers covered by the white-collar
exemptions has been about 8 million.  The increase of 3.6 million
exempt workers in the service sector over this same time represents
about 46 percent of the overall growth in exempt workers.  As a
result of this rapid growth, 29 percent of all exempt workers worked
in the service sector in 1998--up from 19 percent in 1983 (see figure
2). 

   Figure 2:  Percentage of
   Full-Time White-Collar Workers
   Exempt in 1983 and 1998 by
   Industry

   (See figure in printed
   edition.)

Notes:  The percentage estimates represent the average of the high
and low estimates.  See app.  I for a discussion of these estimates.

Service industries included four types of service occupations from
the Current Population Survey (CPS) industry codes:  business and
repair, personal, entertainment and recreation, and professional and
related services.  For definitions of other industries discussed in
this report, see app.  I. 

Source:  CPS Outgoing Rotations Data for 1983 and 1998. 

The demographic composition of the exempt population has
significantly changed in the last 15 years.  In 1998, 42 percent of
exempt workers were women, compared to 33 percent in 1983 (see figure
3).  On the other hand, the gender distribution of nonexempt workers
has not changed in the last 15 years.  About 40 percent of nonexempt
workers were women in both 1983 and in 1998.  These data indicate
that more women than men entered full-time white-collar positions
over this period. 

   Figure 3:  Percentage of
   Full-Time White-Collar Exempt
   and Nonexempt Workforce in 1983
   and 1998 by Gender

   (See figure in printed
   edition.)

Note:  The percentage estimates represent the average of the high and
low estimates.  See app.  I for a discussion of these estimates. 

Source:  CPS Outgoing Rotations Data for 1983 and 1998. 

Full-time workers covered by the white-collar exemptions are much
more likely to work overtime--that is, more than 40 hours per
week--than nonexempt workers (see figure 4).  As figure 4 shows, in
1998, nearly half--44 percent--of the 19 to 26 million full-time
workers covered by the exemptions said they worked overtime at their
primary job.  In 1983, about one-third--35 percent--of full-time
exempt workers worked more than 40 hours per week.  In fact, exempt
workers were more than twice as likely to work overtime in both 1983
and 1998 as nonexempt workers.  In general, the amount of overtime
hours worked by both exempt and nonexempt workers was greater in 1998
than in 1983.  In this regard, in 1998, about 15 percent of exempt
workers worked more than 50 hours per week and 3 percent worked more
than 60 hours per week at their main job.  This compares to 10
percent working more than 50 hours per week and 2 percent working
more than 60 hours in 1983. 

   Figure 4:  Percentage of
   Full-Time Exempt and Nonexempt
   White-Collar Workers Who Worked
   Overtime in 1983 and 1998

   (See figure in printed
   edition.)

Note:  The percentage estimates represent the average of the high and
low estimates.  See app.  I for a discussion of these estimates. 

Source:  CPS Outgoing Rotations Data for 1983 and 1998. 

As figures 5 and 6 show, exempt workers earned substantially more
than nonexempt workers in 1998.  In figure 5, 40 percent of exempt
workers earned $1,000 or more per week, compared to only 7 percent of
nonexempt workers.  Conversely, 57 percent of nonexempt workers
earned less than $500 per week, compared to only 10 percent of exempt
workers. 

Figure 6 illustrates that there are many fewer exempt workers than
nonexempt workers. 

   Figure 5:  Percentage of
   Full-Time Exempt and Nonexempt
   White-Collar Workers in 1998 by
   Weekly Income

   (See figure in printed
   edition.)

Note:  The percentage estimates represent the average of the high and
low estimates.  See app.  I for a discussion of these estimates.  The
zero percentages in this table are the result of rounding and
represent a number between 0 and 0.5 percent. 

Source:  CPS Outgoing Rotations Data for 1998. 

   Figure 6:  Number of Full-Time
   Exempt and Nonexempt
   White-Collar Workers in 1998 by
   Weekly Income (in Millions)

   (See figure in printed
   edition.)

Note:  The numerical estimates represent the average of the high and
low estimates.  See app.  I for a discussion of these estimates.  The
zeros in this table are the result of rounding to the nearest million
and represent a number between 0 and 500,000 people. 

Source:  CPS Outgoing Rotations Data for 1998. 

In 1998, the average weekly earnings of full-time exempt workers were
nearly twice those of nonexempt workers--$1,018\12 weekly compared to
$526 for nonexempt workers.  The difference in earnings between
exempt and nonexempt workers was similar in 1983. 

--------------------
\7 Full-time wage and salary workers exclude self-employed workers
and workers under age 16. 

\8 For each of 257 job titles, DOL provided us with a range
estimate--for example, 10-50 percent--of the employees in that job
category who would probably be exempt.  We arrived at our low
estimate (19 million) by using the lower ends of DOL's individual job
category range estimates, and at the high estimate (26 million) by
using the upper ends of those individual estimates. 

\9 Our work is not an attempt to count the actual number of people
classified as exempt by American employers, but rather to estimate
how many full-time workers are covered by the white-collar
exemptions. 

\10 The number of full-time wage and salary workers grew between 1983
and 1998 as follows:  services, 13 to 24 million; retail trade, 8 to
13 million; manufacturing, 18 to 19 million; finance, insurance, and
real estate, 5 to 7 million; other, 14 to 18 million; and public
sector, 13 to 16 million. 

\11 This estimate and those that follow represent the average of the
high and low estimates.  Please see app.  I for a discussion of these
estimates. 

\12 The earnings figures reported here are earnings from the
respondent's main job before taxes or other deductions including
earnings from overtime pay, commissions, or tips from that job. 

   FEW MAJOR CHANGES IN EXEMPTION
   LAWS AND REGULATIONS SINCE 1954
------------------------------------------------------------ Letter :4

In the 61 years since the enactment of the FLSA, there have been few
major changes to the statutory and regulatory provisions for the
white-collar exemptions.  Between 1938 and 1954, DOL established its
basic set of regulatory tests--the salary-basis test, the
salary-level tests, and the various duties tests.  Although DOL made
a public request for views on restructuring the regulations in 1985,
it has not acted to alter the general way the regulatory tests work
since 1954.  Changes after 1954 have primarily involved adjustments
to the salary-test levels and to the tests applicable to specific
types of workers, such as retail workers, state and local government
employees, and computer programmers. 

In the 16 years following the enactment of the FLSA in 1938, DOL
established the regulatory tests used to determine whether an
employee should be classified as an exempt white-collar worker. 
These tests evolved as DOL's experience with administering the tests
grew.  For example, the first set of regulations in 1938 included a
single test for executives and administrators.  Two years later,
responding to numerous criticisms, DOL drafted two separate
definitions--one for "executives," to apply to people who are bosses,
and another for "administrative" employees, to apply to people who
carry out management policies but who do not supervise other
employees.  DOL made its final change to the structure of the basic
tests in 1954, when it adjusted the exceptions to the salary-basis
requirement. 

Since 1954, statutory and regulatory revisions have, in general,
either (1) adjusted the salary levels upward or (2) modified the
coverage of the exemption, extending or reducing coverage for a
particular type of worker.  The salary levels were adjusted in 1958,
1963, 1970, and 1975.  DOL last attempted to increase these levels in
1981; a Presidential order, however, indefinitely postponed these
increases.  In 1961, statutory and regulatory revisions eliminated a
separate exemption covering most retail workers and specifically
included these workers under the white-collar exemptions.  Other
statutory and regulatory changes expanded coverage of the exemptions
to teachers (1967)\13 and certain higher-wage computer professionals
(1992).  A regulatory revision in 1992 limited the effect of the
salary-basis requirement for state and local governments. 

All statutory and regulatory revisions on white-collar exemptions are
presented in appendix II.  The major changes are summarized in table
3. 

                                Table 3
                
                     Summary of Major Statutory and
                   Regulatory Revisions to the White-
                           Collar Exemptions

Year of revision        Summary of revision
----------------------  ----------------------------------------------
1938 through 1954       Basic regulatory tests set forth in
                        regulations

1961                    Separate retail trade exemption repealed but
                        retail employees were included, with a
                        limitation, under the general coverage of the
                        white-collar exemption

1967                    FLSA was applied to public educational
                        institutions but teachers and school
                        administrators were included under the
                        exemption

1973                    The equal pay provision of the FLSA was made
                        applicable to all those included under the
                        white-collar exemption

1992                    Under certain circumstances, state and local
                        government workers were excepted from selected
                        aspects of the salary-basis requirement

                        Certain computer professionals earning over 6-
                        1/2 times the minimum wage were exempted from
                        the FLSA, even though they were paid an hourly
                        wage
----------------------------------------------------------------------
Source:  GAO analysis of statutory and regulatory provisions. 

--------------------
\13 As pointed out in table 3, public educational institutions were
not covered by the FLSA until 1967. 

   EMPLOYERS BELIEVE THE
   REGULATORY TESTS ARE TOO
   COMPLICATED AND OUTDATED FOR
   THE MODERN WORK PLACE
------------------------------------------------------------ Letter :5

From our reviews of 166 federal court cases\14 involving litigation
on this subject and 66 DOL compliance cases,\15 as well as our
discussions with employers, DOL officials, and various legal and
economic experts, the following three issues stood out as being of
particular concern to employers: 

  -- First, the complex requirements of the salary-basis test or the
     so-called "no-docking" rule presented possibly the greatest
     potential liability for employers and made it difficult to
     account for employees' time and actions. 

  -- Second, the traditional limits of the white-collar exemptions
     between the highly paid, very skilled nonexempt technical
     workers and the exempt professional and administrative employees
     have been blurred in the modern work place. 

  -- Third, the requirement for independent judgment and discretion
     on the part of administrative and professional employees was a
     major area of contention in DOL audits involving the
     white-collar exemptions. 

--------------------
\14 We reviewed 5 years of judicial opinions (1994 through 1998) for
both federal district court and federal appellate court cases. 

\15 We reviewed 66 compliance cases closed in the past 2 years in
four DOL field offices. 

      EMPLOYERS CITE PERCEIVED
      DIFFICULTIES OF SALARY-BASIS
      TEST
---------------------------------------------------------- Letter :5.1

The salary-basis test requires that exempt white-collar employees be
paid a set salary each pay period, rather than an hourly wage.  The
test appears to rest upon the assumption that employers would pay
managerial and professional employees who are key to their business
operations a guaranteed salary regardless of the number of hours
worked.  In the DOL enforcement program, this test is viewed as an
accurate indication of managerial and professional status.  The test,
however, effectively limits the ability of employers to "dock" exempt
employees' pay for such things as part-day personal absences and
disciplinary violations (hence, the so-called no-docking rule). 
Employers object to the test because in their opinion (1) compliance
requires exacting adherence to the no-docking requirements, leaving
them vulnerable to private lawsuits by multiple, well-paid employees;
and (2) it limits their ability to hold their exempt employees
accountable for their time and actions. 

In general, DOL regulations specify that employees can be exempt
executives, administrators, and professionals only if they are paid
on a salary basis--that is, employers must pay them a full salary for
any week worked "without regard to the number of days or hours
worked." Although there are exceptions to this rule,\16

a private employer\17 may not dock an exempt employee's pay for
absences of less than a full day--for whatever reason--without
violating the salary-basis test.  In addition, neither public nor
private employers can dock an exempt employee's pay for periods of
less than a week to enforce disciplinary rules except for a violation
of a safety rule of major significance.  Employers, therefore, must
pay exempt employees a full weekly salary even though the employees
may, for example, take time off during the day for an extended lunch
or a visit to the dentist.  Further, employers cannot suspend exempt
employees without pay for less than 1 week for such things as
tardiness or unexplained absenteeism. 

In our interviews, employers and their representatives discussed the
complex requirements of the salary-basis test, and their concerns
that if they should not comply with the requirements, they may face
lawsuits brought by multiple (and possibly highly paid) employees for
back wages and other damages.  Under the FLSA, employees may sue
their employer either individually or collectively for up to 2, and
in some cases 3, years of back wages plus other damages.  In our
review of federal court cases, we found that the salary-basis test
has been the central focus of cases brought by employees against
their employers.  Our review of 166 federal cases involving the
white-collar exemptions litigated in the past 5 years showed that
about 50 percent involved employee suits alleging that employers
improperly claimed exemptions for employees who did not meet the
salary-basis test. 

For the most part, the salary-basis cases involved groups of
supervisory or professional employees collectively suing their
employers.  However, the large majority of cases were initiated by
groups of managerial public employees (such as police and fire
chiefs, senior corrections officers) against state and local
governments,\18 most often because the government could suspend or
had suspended the pay of exempt employees as a penalty for
disciplinary infractions.  Overall, over 70 percent of the
salary-basis cases were collective actions, and about 70 percent were
lawsuits against state and local governments. 

A 1997 Supreme Court case, Auer v.  Robbins, reduced employers'
potential for liability from these lawsuits.\19 In this case, a large
group of senior police officers sued a city government because its
disciplinary suspensions appeared to apply to exempt officers even
though the city had only suspended one police sergeant.  The court
ruled in favor of the city, and effectively limited employers'
liability for improper pay docking to cases where there is an "actual
practice" or a "significant likelihood" of such action.  Prior to
this decision, some federal cases had ruled that an employer did not
meet the salary-basis test if there were only a possibility that the
employer might improperly dock an exempt employee's pay.  The result
of this ruling is that fewer cases have resulted in liability for
employers--our review of 42 federal cases following Auer showed that
30 were decided in favor of the employers.  However, the full effect
of the Auer ruling has yet to be determined. 

Our discussions with employers and review of other federal cases
showed a variety of circumstances in which employers were uncertain
about the limits of the test.  Some examples\20 of questions
concerning these circumstances include the following: 

  -- What constitutes an "actual practice" of pay docking?  When does
     one instance connote an actual practice? 

  -- When can employers correct instances of improper pay docking and
     not incur liability? 

  -- What happens when an employee has no accrued leave?  Can an
     employer deduct from accrued compensatory time? 

  -- The Family and Medical Leave Act requires employers to give
     employees unpaid leave for serious medical conditions; if such
     leave is taken in partial-day increments, does it violate the
     salary-basis test? 

  -- Under what circumstances can employers pay hourly overtime to
     exempt workers and maintain time sheets or set work hours? 

  -- Can an employee be disciplined for failure to complete a work
     shift? 

  -- An employer cannot suspend an exempt employee for less than a
     week; if the suspension is for more than a week, must the
     employer suspend the employee only in weekly increments? 

In Auer, the Supreme Court expressly deferred to the Secretary of
Labor as the expert on regulatory interpretation of the salary-basis
test.  According to legal experts we talked to, clear guidance to
employers from DOL on the technical requirements of the test may
resolve some uncertainties.  However, they indicated that it can be
difficult for the public to locate published legal advice from DOL. 
As one solution, they suggested that DOL make its "opinion
letters"--legal guidance it routinely provides to individual
employers--more widely available to the general public, for example,
by putting them on the Internet. 

However, even where the requirements of the salary-basis test were
relatively clear, employers argued that the effects on their
operations created anomalies.  In the retail industry, for example,
employers cannot dock an exempt employee's pay to recoup losses from
cash shortages or employee theft.  If they do so, the employee is
considered nonexempt and entitled to overtime wages.  Employers
complained that this rule unduly limits their ability to recover
losses where responsibility clearly rests with one employee, such as
when an employee delivering cash receipts to a bank loses the cash or
an employee uses a corporate credit card for personal items. 

Officials from three large local governments\21 told us that the
salary-basis test made it difficult to penalize employees with
systematic disciplinary steps.  All three local governments are
heavily unionized and to satisfy the union contracts they must use
"progressive discipline." This means that before harsh disciplinary
actions--such as long-term suspensions or termination--can be used to
discipline an employee, less severe disciplinary measures--such as
part-day suspensions without pay--must be taken to alert the employee
to performance problems.  However, the officials contend that the
salary-basis test prevented them from taking such lesser measures. 

DOL officials told us that they believe that payment on a salary
basis remains one of the best indicators of managerial and
professional status.  This belief is longstanding.  As a 1940 DOL
report explained: 

     .  .  .  The term "executive" implies a certain prestige,
     status, and importance.  Employees who qualify under the
     definition are denied the protection of the act.  It must be
     assumed that they enjoy compensatory privileges and this
     presumption will clearly fail if they are not paid a salary
     substantially higher than the wages guaranteed as a mere minimum
     under section 6 of the act.  In no other way can there be
     assurance that section 13 (a) (1) will not invite evasion of
     section 6 and section 7 for large numbers of workers to whom the
     wage-and-hour provisions should apply.  Indeed, if an employer
     states that a particular employee is of sufficient importance to
     his firm to be classified as an "executive" employee and thereby
     exempt from the protection of the act, the best single test of
     the employer's good faith in attributing importance to the
     employee's services is the amount he pays for them.  The
     reasonableness and soundness of this conclusion is sustained by
     the record. 

A 1949 DOL report rejected proposals to eliminate the salary-basis
test, commenting that "[C]ompensation on a salary basis appears to
have been almost universally recognized as the only method of payment
consistent with the status of the `bona fide' executive .  .  . 
[and] is one of the recognized attributes of administrative and
professional employment." These principles were again reaffirmed in
the report and recommendations of a 1958 hearing report on proposed
revisions to the regulations. 

Compliance investigators in DOL's Wage and Hour Division said that
the salary-basis test is a key enforcement test.  Investigators told
us that the first review they routinely undertake involves
determining whether the exempt employees are paid on a salary basis. 
However, as one investigator explained to us, testing for compliance
can be difficult because most often exempt employees have worked in
excess of 40 hours and have been paid for at least 40 hours, making
it hard to prove that employers have made improper deductions to the
salaries of exempt workers.  Although we found that the salary-basis
test was not the central issue in most compliance cases we reviewed,
it was critical to some cases.  In one case, for example, a gas
service station owner paid managers and assistant managers a salary
of $400 per week and required them to work 60 hours per week. 
Although the owner claimed the employees to be exempt executives, the
investigator successfully challenged their executive status because
the owner reduced their pay if they worked less than 60 hours per
week. 

--------------------
\16 Exceptions to the rule include deductions of a day or more for
personal reasons other than sickness and accident, deductions of a
day or more for sickness if in accordance with company policy, and
deductions for violation of safety rules of major significance. 

\17 A 1992 amendment to the regulations allowed state and local
governments to dock employees' pay for part-day absences under
certain circumstances. 

\18 In a recent decision, Alden v.  Maine, 119 S.Ct.  2240 (1990),
the Supreme Court ruled that state employees cannot sue a state in
state courts for overtime wages under the FLSA if the state has not
waived its sovereign immunity.  A previous Supreme Court decision,
Seminole Tribe of Fla.  v.  Florida, 517 U.S.  44 (1996), precluded
such suits in federal courts. 

\19 519 U.S.  452. 

\20 In commenting on this report, DOL stated that it believes some of
the legal issues raised by employers have been authoritatively
addressed in either the case law or the law itself.  Our review of
the case law indicates that legal conclusions on issues related to
the salary-basis test can vary depending upon the factual
circumstances of individual cases. 

\21 Because it is clear that local governments (and other employers)
violate the salary-basis test when they discipline their management
officials by suspensions without pay, all three local governments we
talked to have made large groups of their employees nonexempt rather
than give up their disciplinary systems.  One now pays its senior
police officials overtime, another has limited its exempt workers to
those paid over $73,000 per year, and the third has made nearly all
union workers nonexempt (90 percent of its employees are unionized). 

      EMPLOYERS SAY SOME
      PRODUCTION WORKERS ARE
      EQUIVALENT TO WHITE-COLLAR
      EMPLOYEES
---------------------------------------------------------- Letter :5.2

Nonsupervisory employees may be exempt from the FLSA if they can be
classified as either administrative or professional employees. 
However, the administrative exemption is limited to those employees
who perform nonmanual (or nonproduction) work "directly related to
the management policies or general business operations" using
independent judgment and discretion, and the professional exemption
is generally limited to occupations requiring advanced academic
degrees.\22 Thus, as interpreted for the past 60 years, these
exemptions do not apply to many technical workers--nonsupervisory
line workers who work to produce the employer's goods and who do not
have the requisite academic degree for a recognized profession in
which they are employed. 

In our discussions with employers and state and local government
representatives, both groups argued that the traditional limits of
the white-collar exemptions are outdated in the modern work place. 
They believed that certain highly skilled, well-paid line workers
should be treated as exempt workers because they have the knowledge
equivalent to an exempt professional.  Officials from manufacturing
employers pointed to new technology used in factory work places,
which they said required advanced technical skills but required far
less traditional "manual" labor.  Moreover, they told us that while
these workers may have to follow precise written guidelines to
perform their work, prescribed procedures were key to modern quality
control.  State and local representatives pointed to job
classifications within their organizations which involved line work
but which required the knowledge and experience of a civil service
professional. 

To illustrate the point, one company official described the job of
technicians who maintain unmanned factories around the country.  In
her company, one technician, relying upon standardized instruments,
monitors each automated factory.  The official compared the nonexempt
line jobs of these well-paid (about $70,000 per year) technicians
with those of her company's engineers.  Both groups held similar jobs
and earned comparable pay, but because the engineers had professional
degrees and the technicians did not, only the engineers could be
classified as exempt professional employees. 

In general, employers pointed to the differences between exempt and
nonexempt employee status as creating difficulties in managing the
workforce, particularly in what they referred to as crossover
positions like the technicians described above.  According to the
employers, nonexempt employees have less flexibility in work
shifts--any work over 40 hours must be paid at overtime rates, even
if the employee is planning on working less than 40 hours in the next
work-week.  Further, employers claim that workers look on exempt
status as prestigious, allowing greater possibility of management
promotion.  In addition, employers also believed that adherence to
strict written guidelines--one major distinction between exempt and
nonexempt workers--is necessary in a modern, efficient work place. 

In recent years, the distinctions between production and
nonproduction workers, and between professional and technical
production workers, have been increasingly blurred.  The federal
court cases and DOL compliance cases we reviewed provided
illustrations of recent distinctions made between exempt and
nonexempt nonsupervisory workers.  These cases included the
following: 

  -- A large publishing firm employed about 50 production editors,
     each of whom managed the final publication processing of the
     company's books.  One of these production editors sued the
     company, claiming that she was improperly classified as an
     exempt employee.  A federal appellate court ruled that, although
     her job involved production work, she was an exempt employee. 
     It found that her work was a major assignment of the company
     and, therefore, that it was directly related to the management
     policies or business operations of the company.\23

  -- An insurance company employed automobile damage appraisers to
     determine the amount to be paid on auto insurance claims.  A
     federal district court found the appraisers to be nonexempt
     production workers because they did not do work related to the
     management policies or business operations of the firm.  Rather
     than administratively running the business, they carried out the
     daily affairs of the company.\24

  -- A title insurance company used escrow closers to conduct final
     property settlements.  A federal district court held that these
     were nonexempt employees who were carrying out the day-to-day
     operations of the company\25

and that it was irrelevant that they were not part of the company's
production department. 

  -- One southern state's Department of Agriculture employed dairy
     inspectors and food safety inspectors.  The dairy inspectors had
     academic training and experience in the specific area of dairy
     or animal sciences, while the food safety inspectors had more
     generalized academic training in biological sciences.  DOL
     compliance review determined that only the dairy inspectors met
     the educational requirement of the professional exemption. 

--------------------
\22 There are, however, exceptions for certain specific occupations
such as computer programming, based on a 1990 congressional
enactment, and work of an artistic nature. 

\23 Shaw v.  Prentice Hall Computer Publishing, Inc., 151 F.3d 640
(7\th Cir.  1998). 

\24 Reich v.  American International Adjustment Co., Inc., 902 F. 
Supp.  321 (D.  Conn.  1994). 

\25 Reich v.  Chicago Title Insurance Co., 853 F.  Supp.  1325 (D. 
Kan.  1994). 

      THE REQUIREMENT FOR
      INDEPENDENT JUDGMENT AND
      DISCRETION IS DIFFICULT TO
      APPLY
---------------------------------------------------------- Letter :5.3

To be classified as either an exempt administrative or professional
employee, each employee must exercise independent judgment and
discretion in carrying out his or her job duties.  In general, the
requirement for exercising independent judgment and discretion means
that employees have the freedom to make choices about matters of
significance to their employers, without immediate supervision or
detailed guidelines.  Factors which are taken into account include
(1) the amount of supervision, (2) the amount of written guidelines,
and (3) whether the work involves routine matters.  For example, a
newly hired accountant may be given work that is closely supervised,
involving rote work with set procedures.  In that case, the
accountant would be nonexempt, even though he or she may have full
professional certification. 

Our discussions with employers and DOL investigators indicated that
this aspect of the regulations is particularly difficult to apply for
both the employers and the investigators.  Employers complained that
the standards for the independent judgment requirement were confusing
and applied in an inconsistent manner by DOL.  Thus, employers were
unsure of how to properly classify administrative personnel. 
According to DOL investigators, determinations about independent
judgment and discretion can be the most difficult part of a
compliance review.  To assess this requirement, an investigator must
review both the general duties of the position and the specific
duties of the employee.  Further, the determination may hinge upon
how an individual employee views his or her own duties.  For
instance, one administrative assistant may look at his job as
answering telephone calls and following orders, while another person
in the same position might describe the job as involving the
independence to establish office procedures and respond to incoming
client inquiries. 

The compliance cases we reviewed included a number of instances where
the standard for independent judgment and discretion was key to
determining the employee's status.  Situations where this question
arose included the following: 

  -- A trucking company employed dispatchers to organize and schedule
     truck routes.  Two of the dispatchers negotiated with other
     companies to obtain contracts for truck loads, as well as
     scheduling truck routes.  The DOL compliance investigator
     allowed the administrative exemption for the two senior
     dispatchers, but not for the other dispatchers. 

  -- A firm provided library services to professional firms.  Firm
     officials claimed that certain librarians were exempt as either
     administrative or professional employees.  The DOL investigator
     disagreed, finding the librarians were nonexempt because their
     work--filing and updating loose-leaf volumes--was "routine and
     not dependent on a professional degree."

  -- An architectural firm hired professional architects and
     engineers.  The firm classified its employees according to
     experience, but considered them all exempt.  The DOL
     investigator found that architects and engineers at the entry
     level were nonexempt, because they did not exercise discretion
     and independent judgment in their jobs. 

In our review, we noted certain cases in which an employer conducted
its own self-audit and used this to negotiate a final settlement with
DOL.  One large accounting firm agreed to conduct a self-audit of all
of its entry-level tax reporting accountants, and DOL agreed to not
question the application of the professional exemptions to second-
and third-year accountants.  In another example, DOL investigators
found certain job titles at a commercial bank that appeared to
involve nonexempt work.  To settle the compliance questions, the bank
hired a law firm to conduct an audit of individual employee
classifications.  The lawyers examined the bank's FLSA
classifications and recommended that changes be made.  With the
approval of DOL, the bank accepted the reclassifications suggested by
the audit and agreed to pay the necessary back wages. 

   EMPLOYEES SAY THAT INFLATION
   AND OVERSIMPLIFICATION HAVE
   UNDERMINED EXEMPTION LIMITS
------------------------------------------------------------ Letter :6

Employee representatives and other experts were particularly
concerned that the use of the exemptions be limited, maintaining the
40-hour work-week standard for as many employees as possible.  To do
this, they were of the opinion that the regulatory tests should
provide the type of protection originally intended.  In this regard,
the following two issues seemed particularly important to employees: 

  -- The salary-test levels that underpin the regulatory framework
     have been unchanged since 1975.  Because of inflation, the
     current salary-test levels are now near the minimum-wage level,
     rendering the application of certain regulatory tests to the
     current workforce virtually meaningless. 

  -- The duties test that determines who can be classified as an
     exempt executive has been increasingly simplified by judicial
     opinions.  When combined with the low salary-test levels,
     employees believe that few protections remain for lower-income
     supervisors. 

      INFLATION HAS EFFECTIVELY
      ELIMINATED IMPORTANT ASPECTS
      OF THE REGULATORY TESTS
---------------------------------------------------------- Letter :6.1

In determining whether an employee is exempt from the FLSA as an
executive, administrator, or professional, the first consideration is
the employee's salary.  Since 1949, employees have been divided into
three groups according to their weekly earnings.  As described
earlier in table 2, the standards vary depending on whether an
employee is to be classified as an executive, administrator, or
professional.  For the executive exemption, the three groups
currently are as follows: 

  -- Employees earning less than $155 per week are automatically
     nonexempt (or subject to the FLSA requirements). 

  -- Employees making at least $155 but less than $250 per week are
     nonexempt unless their duties meet the rigorous standards of the
     so-called long duties test.  The long duties test for executives
     requires that employees' duties include such things as the
     authority to hire or fire other employees and the ability to
     exercise discretion; most significantly, though, it sets
     percentage limitations on the amount of nonmanagerial work an
     exempt employee can perform in a work-week.  Specifically,
     employees qualify as exempt employees only if no more than 20
     percent (or less than 40 percent for retail and service
     employees) of their jobs involve nonmanagerial or
     nonprofessional work. 

  -- Employees earning at least $250 per week are exempt as long as
     their duties meet the less strict standards of the so-called
     short duties test, an abbreviated version of the long duties
     test.  The short duties test does not specify percentage
     limitations on the amount of nonmanagerial work an exempt
     employee can perform.  For executives, the test is limited to
     requiring that employees supervise two or more workers and that
     their primary duty is managerial. 

The practical differences between the long and short duties tests are
significant.  To illustrate, consider a cook who supervises a crew of
other workers.  If the cook's salary was $200 per week, and he was
thus subject to the long duties test, he would be nonexempt (and
entitled to overtime wages) if he spent more than 40 percent of his
time on nonmanagerial tasks--work such as cooking food or cleaning
the kitchen.  If, however, the cook's salary was $250 per week, the
cook would be an exempt executive as long as his primary duty was
management and included the customary and regular direction of at
least two other employees, even though he may spend a lot of time
cooking food or cleaning the kitchen. 

The salary levels used to determine into which of the three salary
categories an employee falls have not been changed since 1975. 
During that time period, salaries in the nation have risen
considerably.  As a result, the salary levels used by DOL for this
purpose, which in 1975 were considered fairly high, are now below the
level of the federal minimum wage in the instances of the base salary
levels.  Even the higher level, $250 per week, is equivalent to an
hourly wage that is only $1.10 per hour higher than the current
minimum federal hourly wage of $5.15 for a 40-hour work-week.  For
the salary levels used in this determination to represent the same
level of purchasing power now as they did in 1975, they would need to
be considerably higher than their current levels.  For example, the
highest of levels, $250 per week, would have to be $757 per week, for
an annual salary of about $39,400.  In figure 7, we examine the
highest salary level over the 49-year period between 1949 and 1998,
and we compare the actual level included in the regulations with the
level necessary to keep pace with inflation. 

   Figure 7:  Actual and
   Inflation-Adjusted Highest
   Salary Test, or Upset Test, for
   Weekly Income, 1949-1998

   (See figure in printed
   edition.)

Note:  Upset test numbers are adjusted for inflation using the
Consumer Price Index for all Urban Consumers (CPI-U), with 1975 as
the base year. 

Source:  Data for the actual upset test are from 29 C.F.R.  chap.  V,
part 541; inflation-adjusted upset test calculated by GAO. 

To see the effect of inflation on the application of the regulatory
tests, consider again the example of the supervisory cook.  Today,
the cook would be automatically nonexempt only if his salary was less
than $8,060 per year--the equivalent of $3.88 per hour, $1.27 less
than the current federal minimum wage.  The strict long duties test
would apply only if he made less than $13,000 (equivalent to an
hourly rate not much higher than the minimum wage for a 40-hour
work-week).  And, as long as he made $13,000, he would be presumed to
be an executive if his primary duty was management--for example, if
he can hire and fire workers. 

However, if the cook's salary was adjusted to include the inflation
occurring between 1975 and 1998, the application of the regulatory
tests would be very different.  Using salary figures adjusted for
inflation, the cook would be automatically nonexempt as long as he
earned less than $24,400.  If he earned between about $24,400 and
$39,400, he would be exempt only if his work met the long duties
test.  And, if he earned more than $39,400, he would be an exempt
executive if his primary duty were management. 

Because of inflation, the percentage of full-time workers who
potentially fell into each of the three salary level categories was
far different in 1975 than in 1998\26 --specifically,

  -- In 1975, about 30 percent of the full-time workforce would have
     been automatically nonexempt workers; in 1998, only 1 percent of
     the full-time workforce were automatically nonexempt.\27

  -- In 1975, about 30 percent of the full-time workforce would have
     been nonexempt unless they met the percentage limitation on
     performing nonexempt work; in 1998, the long duties test would
     apply to only 8 percent of the full-time workforce. 

  -- In 1975, about 40 percent of the full-time workers could have
     qualified as exempt workers with the application of the short
     duties test; in 1998, 91 percent of the workers were under the
     short duties test. 

--------------------
\26 The data we used for these estimates include full-time wage
workers as well as full-time salaried workers age 16 years and over. 
The data were provided by BLS and are unpublished tabulations from
the CPS.  The data for 1998 are annual averages while the data for
1975 are for the month of May, when annual averages were not
available.  Self-employed workers are excluded. 

\27 These percentages are approximate; the CPS data provide the
percentage of workers who earn under $150 per week, rather than under
$155 per week, the base salary for exempt executive and
administrative employees. 

      EMPLOYEES SAY THAT DUTIES
      TEST OFFERS LITTLE
      PROTECTION FOR LOWER-INCOME
      SUPERVISORS
---------------------------------------------------------- Letter :6.2

During the past 20 years, it has become increasingly easy to classify
a supervisory employee as an exempt executive.  If an employee makes
over $250 per week ($13,000 per year), the employee may be an exempt
executive if he or she meets two criteria.  First, the employee must
customarily and regularly direct the work of two or more employees;
and second, the employee's primary duty must involve management. 
Unlike the administrative and professional exemptions, there is no
express requirement for independent judgment and discretion. 
However, the regulations specify that, "as a rule of thumb," an
employee who spends more than 50 percent of his or her time on
management tasks would have management as a primary duty. 

Two federal court decisions in 1982 clarified the test for
determining whether an employee earning over $250 per week has
management as a primary duty.  The cases, involving litigation
between DOL and the Burger King Corporation,\28

applied the executive exemption to assistant managers at the
fast-food restaurants.  First, one decision held that the duties of
Burger King assistant managers were primarily managerial, even though
the company provided them detailed instructions on how they were to
perform their work.  Second, both courts found that the 50-percent
"rule of thumb" limitation on nonmanagerial work was only one factor
to consider when determining employees' primary duty, and that their
managerial duties could be carried out at the same time they were
performing manual work.  Thus, assistant managers could be exempt
executives even if they spent most of the day cooking hamburgers--as
long as they were in charge of the restaurants during their shifts. 

While employers appreciate the simplicity and the clarity of the
executive duties test, union representatives complained that the
judicial rulings following the Burger King decisions have
oversimplified the executive test.  Under the regulations, as
currently interpreted, almost any employee who is assigned to
supervise two or more employees in a particular "department" of the
company can be classified as an exempt employee.  According to the
union officials, employers have adjusted their work places to include
many new levels of supervision in order to create exempt executive
positions.  Thus, where a grocery store originally had one or two
store managers, it now has many different departments--the meat
department, the produce department, and others--headed by exempt
executives. 

Federal case law in the 5-year period from 1994 through 1998 included
hardly any instances in which a court overturned an employer's
classification of a lower-income supervisor as an exempt executive. 
In the 32 cases we identified as relating to the executive exemption
duties test, about one-third (12 cases) involved employees whose
salaries were $500 per week (or $26,000 per year) or less.  Of these
12 cases, only 1 resulted in a favorable ruling for the employee. 
These cases included a wide range of employees, such as: 

  -- an aquatics director of a community swimming pool paid $376 per
     week,

  -- a produce department manager of a grocery store paid $450 per
     week,

  -- a dietary manager of a nursing home paid $341 per week, and

  -- a loss-prevention manager of a department store paid $423 per
     week. 

All of these employees claimed that their jobs consisted primarily of
nonmanagerial work--for example, life guarding, stocking shelves, and
cooking.  Despite evidence of large proportions of nonmanagerial
work, the courts found all but one employee to be exempt executives. 

Discussions with DOL investigators and attorneys suggested that the
Burger King decisions and the low salary-test levels have had a major
effect on their investigation of cases involving exempt executives. 
Since the decisions, their policy manual has been revised to require
that investigators consider percentage limitations as only one factor
when assessing the employee's primary duty.  One attorney noted that
in recent years little litigation had been initiated by DOL over the
executive status of supervisory employees.  He indicated that,
although there may be situations in which the exempt executive
classification of an employee supervising two or more workers could
be challenged, those situations are very limited. 

--------------------
\28 Donovan v.  Burger King Corp., 672 F.2d 221 (1\st Cir.  1982),
and Donovan v.  Burger King Corp., 675 F.2d 516 (2d Cir.  1982). 

   CONFLICTING INTERESTS OF
   EMPLOYERS AND EMPLOYEES MAKE
   RESOLUTION OF CONCERNS
   DIFFICULT
------------------------------------------------------------ Letter :7

Legal and economic experts have proposed various ways to deal with
the concerns raised by employers and employees, ranging from
tinkering with particular provisions of the regulations to a major
overhaul of the FLSA.  However, proposals to change the present law
or regulations all affect the regulatory balance between the desires
of employers and those of employees, and there are competing
interests that must be carefully weighed before any changes can be
made.  For a number of years, DOL has been reluctant to alter the
existing tests in view of these competing interests.  To illustrate
some of these considerations, we discuss four proposals that have
been made by experts to revise the current regulations, and summarize
the general views of employers and employees on each. 

      ELIMINATE THE SALARY-BASIS
      TEST
---------------------------------------------------------- Letter :7.1

From the employer's point of view, the salary-basis test presents
complex regulatory requirements--for example, the limitations on the
use of pay suspensions to sanction employees' actions--that have
nothing to do with managerial or professional status but that can be
the source of potential legal liability for unwary employers. 
However, if the test were eliminated, the exemption could be applied
to both hourly-wage and salaried workers and only the other two
tests--the salary levels and the duties test--would remain as
indicators of managerial and professional status.  DOL contends that
salary remains the general method of compensation for key managerial
and professional employees.  From the standpoint of employees, the
salary-basis test is a key enforcement tool to protect workers from
employers who do not comply with the law because it is an objective
measure of managerial and professional status.  For employees, it is
particularly important today because the salary-test levels are so
low, and without the salary-basis test, DOL would be left only with
the difficult-to-apply duties test as the single test of employer
compliance. 

In commenting on this report, DOL further explained the rationale for
the salary tests.  According to DOL, the salary tests have been an
integral part of the definitions for the exemptions since 1940.  It
believes that the statutory terms "executive, administrative, or
professional" imply a certain prestige, status, and importance, and
an employee's salary serves as one indicator of his or her status in
management or the recognized professions.  It is an index that
distinguishes the bona fide executive from the working squad leader,
or distinguishes the clerk or technician from one who performs true
administrative or professional duties.  DOL said that salary remains
a good indicator of the degree of importance attached to a particular
employee's job, which provides a practical guide, particularly in
borderline cases, for distinguishing bona fide executive,
administrative, and professional employees from those who were not
intended by the Congress to come within the categories of this
exemption.  In its years of experience in administering the
regulations, DOL said it has found no satisfactory substitute for the
salary test.  The arguments that the salary test is not needed or
that it does not help draw the proper line between exempt and
nonexempt employees have been offered before, including as part of
the hearings and deliberations over amending the regulations in 1940,
1949, and 1958.  As then, the DOL sees no new or more valid reasons
that have been offered for eliminating the salary test from the
regulations than were considered as part of the previous hearings on
this question. 

      RAISE THE SALARY-TEST LEVELS
---------------------------------------------------------- Letter :7.2

If the salary-test levels were raised, the regulatory structure could
incorporate the effects of inflation and function as originally
envisioned.  With a higher salary test, both the long and short
duties tests would be applied once again to segments of potentially
exempt workers.  Nearly everyone we talked to--employers, employees,
and experts--agreed that the current salary-test levels are too low
and should be increased to higher, more reasonable levels.  However,
they disagreed sharply on whether the duties tests should remain the
same after the salary-test levels were raised. 

Employers, particularly retail employers, were opposed to reviving
the long duties test and, with it, the percentage limitation on the
amount of nonmanagerial work that an exempt employee would be allowed
to do.  Retail employers argued that the percentage limitations are
outdated in modern store management.  They said that, in recent
decades, store management has changed dramatically.  In the 1960s,
stores were open for far fewer hours than today and employed one
store manager and a full-time workforce.  Now, stores are open as
much as 24 hours per day, and many part-time workers and managers
work each shift.  Under these conditions, employers contended that
managers must pitch in and work the cash register or stock the
shelves, while at the same time managing the store operations. 

Union representatives, on the other hand, believe that the long
duties test, with its percentage limitations, contains critical
criteria for assessing managerial and professional status.  They
contend that the amount of nonexempt work is a basic indicator of
managerial status.  They argue that if a worker is engaged in
primarily nonexempt work, the worker should be classified as a
nonexempt employee, regardless of how his or her employer categorizes
the employee's position.  Moreover, they argue that the long duties
test still allows retail employers to designate the manager in charge
of a store as exempt, notwithstanding how much nonexempt work he or
she performs, under the so-called "sole charge" exception to the
requirement for a limit on nonexempt work. 

      ADD A CATEGORY OF KNOWLEDGE
      WORKER TO THE EXEMPTIONS
---------------------------------------------------------- Letter :7.3

This proposal would add a new category of exemption--the knowledge
worker--to the executive, administrative, and professional
exemptions.  This category would include well-paid, highly skilled,
nonsupervisory workers who are currently not covered by the
exemptions.  It would require a new definition of exempt employees,
with new criteria and separate salary-test levels. 

Employers believe that the current exemptions leave a gap by not
including workers who are not engaged in traditional manual labor but
who follow detailed procedures to perform their jobs.  Because these
workers can be both highly skilled and well paid, employers think
that they should be classified similarly to exempt professionals. 
This would allow both the worker and management more flexibility in
scheduling and offer the worker what employers consider prestigious
jobs in management positions. 

Employee representatives argue against the expansion of the
exemptions to include these technical workers.  They believe that
today's computer-assisted technicians are the modern equivalents of
traditional factory workers, and the employee representatives said
that there always have been workers who could have been classified as
knowledge workers.  They think that the same principles underlying
the historical limitations on work hours and requirements for
overtime pay should apply to the modern workforce.  They assert that
while everyone would like more flexibility in the work place, in
reality exempt status means that employees work longer hours for less
pay.  And, although they agreed that there are some workers who view
working longer hours as a way to management promotion, they think
that the majority of workers still would like to have restrictions on
the number of hours they can be required to work without additional
compensation.  Finally, they told us that discretion and independent
judgment remain the key indicators of professional and managerial
status--rote work, however well paid, shows that a worker is only a
"cog in the wheel," not a key managerial employee. 

      ADJUST SALARY LEVELS AND
      DUTIES, APPLYING AN INCOME
      CEILING TO NONEXEMPT STATUS
---------------------------------------------------------- Letter :7.4

Finally, another proposal would adjust the salary-basis test, the
salary-test levels, the duties tests, and the categories of workers
in a new regulatory framework.  One basic framework would be similar
to that already in the regulations--there would be an income floor
below which all workers would be nonexempt, combined with an income
ceiling above which workers would be presumed exempt. 

As we noted above, although there is nearly universal agreement that
the salary levels should be raised, adding an income ceiling is much
more controversial.  Employers and employees disagree on whether
there should be a ceiling, and if so, how high it should be, and
what, if any, duties tests should apply to the different income
levels. 

Issues regarding the ceiling income level include the geographic and
industry differences in average salary levels and the possibility of
indexing the salary level to increases in average compensation.  The
questions related to the duties test involve whether there should be
any duties test at all above the ceiling--if not, there would be a
conclusive presumption that employees with incomes higher than the
ceiling are exempt managerial or professional employees. 
Alternatively, if some duties test were retained, an employee's
exemption could be challenged if his or her duties did not include
sufficient managerial or professional responsibilities. 

For employers, adding an income ceiling would bring more certainty
into the classification of higher-paid workers.  Depending on how the
duties test for those above the income ceiling was applied, a ceiling
could reduce employers' potential liability for violations of the
salary-basis test by eliminating the need for the test among the
highest-paid workers. 

For employees, assuming that there was no applicable duties test, an
income ceiling would eliminate the requirement for a 40-hour
work-week for higher paid workers.  Union representatives believe
that, in effect, it would penalize workers for being relatively
highly paid.  As our data indicated, exempt workers are more likely
than nonexempt workers to work more than 40 hours.  Employee
representatives suggest that the increase in hours of work for exempt
employees presents special challenges for the increasing numbers of
dual-earner households with middle or upper-middle levels of income. 
They argue that regardless of income level, it continues to be
important that workers have control over both the number and timing
of work hours to better balance the competing demands of work and
family. 

These proposals show how the divergent interests of employers and
employees require tradeoffs to resolve the underlying issues with the
white-collar exemptions. 

   CONCLUSIONS
------------------------------------------------------------ Letter :8

The concerns of employers and employees about the operation of the
white-collar exemptions in today's work place involve all aspects of
the regulatory tests--the salary-basis test, the salary-test levels,
and the duties requirements.  DOL has not updated these tests in
decades, with the result that the salary-level tests are virtually
meaningless and there are mounting complaints by employers about
ambiguity in the requirements.  Although DOL made some efforts to
revise the tests in the 1980s, it has not acted because of the
difficulty of getting consensus on the changes.  Recent attempts to
correct the regulatory tests for specific groups--state and local
governments and computer professionals--have done little to alleviate
the general problems. 

Given the economic changes in the 60 years since the passage of the
FLSA, it is increasingly important to readjust these tests to meet
the needs of the modern work place.  However, the different
regulatory tests interact with one another, and a change to one test
can undermine or strengthen the operation of other regulatory
provisions.  For example, elimination of the salary-basis test could
further weaken the protection offered to the lower-income supervisor
who is required to work a 60-hour work-week.  Raising the salary
levels, on the other hand, adds more complexity to the regulatory
tests for executive employees by making the long duties test
applicable to at least some workers.  Resolution of these concerns
requires a careful balancing of the needs of the employers for clear
and unambiguous regulatory standards with those of employees for fair
and equitable treatment in the work place. 

   RECOMMENDATION
------------------------------------------------------------ Letter :9

We recommend that the Secretary of Labor comprehensively review the
regulations for the white-collar exemptions and make necessary
changes to better meet the needs of both employers and employees in
the modern work place.  Some key areas of review are (1) the salary
levels used to trigger the regulatory tests, and (2) the categories
of employees covered by the exemptions. 

   AGENCY COMMENTS
----------------------------------------------------------- Letter :10

We provided a draft of this report to the Department of Labor for its
review and comments.  In its comments, DOL stated that the report was
well balanced and accurate in presenting the issues addressed.  With
respect to the report's recommendation, DOL noted that the
white-collar exemption regulations are on its agenda to be reviewed
in the future.  The Department noted that any change in the current
regulatory structure requires balancing the diametrically opposed,
conflicting interests of the many and differing constituencies that
would be affected, and that the views of interested parties are
intractably held on opposite sides of the various issues under these
regulations.  DOL added that, given the current regulatory
environment, the prospects that it could successfully implement
consensus changes as we recommended are greatly diminished.  In
addition, the agency provided other technical comments, which we
incorporated in the report as appropriate.  (See appendix III for a
copy of DOL's written comments.)

We are sending copies of this report to the Honorable Alexis M. 
Herman, Secretary of Labor; the Honorable Bernard E.  Anderson,
Assistant Secretary for Employment Standards; the Honorable Katherine
G.  Abraham, Commissioner of the Bureau of Labor Statistics;
appropriate congressional committees; and other interested parties. 

Please call me or Larry Horinko at (202) 512-7001 if you or your
staff have any questions about this report.  Other major contributors
to this report are listed in appendix IV. 

Cynthia M.  Fagnoni
Director, Education, Workforce, and
 Income Security Issues

SCOPE AND METHODOLOGY
=========================================================== Appendix I

We used a variety of data sources to develop and provide updated
information on the white-collar exemptions.  To estimate the number
and demographic characteristics of the portion of the American
workforce who were executive, administrative, and professional
employees covered by the white-collar exemptions in 1998 and 1983, we
used the Bureau of the Census' Current Population Survey (CPS)
Outgoing Rotations data.  To identify statutory and regulatory
changes since 1938, we reviewed various legal reports and
publications.  To determine the major concerns of employers and
employees regarding the white-collar exemptions, we reviewed 166
federal court cases on this subject and 66 Department of Labor (DOL)
compliance cases, and held discussions with employers, unions, DOL
officials, and various legal and economic experts.  We performed our
work between November 1998 and June 1999 in accordance with generally
accepted government auditing standards. 

   ESTIMATING THE NUMBER AND
   DEMOGRAPHICS OF WORKERS COVERED
   BY THE WHITE-COLLAR EXEMPTIONS
--------------------------------------------------------- Appendix I:1

To estimate the number and demographic characteristics of exempt
white-collar workers, we did not attempt to count the number of
employees actually classified as exempt by employers but, rather, to
determine how many employees were covered by the regulatory tests for
the white-collar exemptions in 1983 and 1998.  We reviewed several
Bureau of Labor Statistics (BLS) and DOL reports to determine whether
any data sources could be used for our purposes.  After discussions
with DOL and experts, we decided that the CPS Outgoing Rotations was
the best available data source to estimate both the proportion of the
labor force that is covered by the white-collar exemptions and the
demographic characteristics of this population.  The CPS is conducted
by the Bureau of the Census for the BLS.  The Outgoing Rotations data
are collected as part of the basic CPS labor-force interview and
provide data on weekly earnings along with the demographic
information such as age, gender, and race.  The CPS is a survey of
households collected each month from a probability sample of
approximately 50,000 households.  The CPS provides self-reported
information on the labor-force status of the civilian
noninstitutional population 16 years of age and over.  Our analysis
covered calendar years 1998 and 1983. 

      GENERAL METHODOLOGY
------------------------------------------------------- Appendix I:1.1

To estimate the number of workers covered by the white-collar
exemptions using the CPS data, our primary focus was on two
questions:  (1) Was the worker paid a salary?  (2) What was the
worker's primary job classification?  The CPS data include 905
occupational classifications.  Respondents are asked to describe the
kind of work they are doing, and CPS coders classify these duties
into one of the 905 job titles.  To determine which of the 905 job
titles would likely include exempt white-collar workers, we asked the
DOL officials to assess the likelihood of exemption for each
occupation.  DOL officials responded by classifying each of the 905
occupations individually by its likelihood of exemption.  Overall,
DOL officials determined that 257 of the 905 job titles would likely
include exempt workers.  For each of these 257 job titles, DOL
officials provided us with one of four ranges of likelihood that
workers would be covered by the white-collar exemptions:  90-100
percent, 50-90 percent, 10-50 percent, and 0-10 percent.\29

To develop our estimate, we analyzed each of the 257 job titles
likely to include exempt workers.  In determining which of the
workers in the sample would likely be exempt and therefore included
in our estimate, we applied the percentage ranges provided by the
officials at DOL.  To better refine the application of these
percentage ranges (which were estimates of the likelihood that the
positions would include managerial or professional duties), we made
the following assumption:  duties that make an employee more likely
to be covered by the white-collar exemptions are duties that,
generally speaking, elicit a higher salary.  Under this assumption,
as workers have more exempt duties and responsibilities, their
incomes increase--as does the likelihood of being exempt.  Therefore,
we applied the percentage ranges provided by DOL officials to give
increasing weight to workers with higher incomes.  All workers,
except physicians, lawyers, and teachers, earning less than $250 per
week were considered nonexempt and were eliminated from our
calculation of the exempt population.  The exemption for physicians,
lawyers, and teachers does not depend on the income of the employee. 

Using this method for each of the 257 occupations, we generated our
estimates for the low and high estimates of the potentially exempt
population for both 1983 and 1998.  After we estimated the population
covered by the white-collar exemptions, we determined its demographic
characteristics.  Our analysis included information on gender,
industry, weekly earnings, and overtime hours worked. 

Our work presents data for six industry groupings:  (1) services; (2)
retail trade; (3) manufacturing; (4) finance, insurance, and real
estate; (5) public sector; and (6) other.  We developed these groups
by combining 932 detailed CPS industry codes.  The service sector
includes four types of service occupations:  business and repair,
entertainment and recreation, professional and related services, and
personal services.  Retail trade includes all types of retail stores. 
Manufacturing includes durable and nondurable goods.  The finance,
insurance, and real estate category includes banking, credit
agencies, security companies, and insurance and real estate offices. 
Public sector includes federal, state, and local government workers. 
The final category, other industries, includes agriculture, forestry,
fishery, mining, construction, wholesale trade, transportation,
communications, public utilities, and public administration. 

--------------------
\29 These ranges were not overlapping since each occupation is
assigned a different range. 

      DATA LIMITATIONS RELATED TO
      THE USE OF THE CPS
------------------------------------------------------- Appendix I:1.2

There are two major limitations on the use of CPS data.  First, the
CPS occupational classifications do not distinguish between
supervisory and nonsupervisory employees, which is important for the
long and short duties tests under the Fair Labor Standards Act
(FLSA).  Therefore, one job title, "managers and administrators,"
could include the President of General Motors, but it may also
include an office assistant.  Second, CPS respondents self-identify
their duties and some may tend to exaggerate them.  This may result
in overestimates of the number of management employees and,
consequently, may overestimate the number of exempt employees. 

We corrected for these data limitations by giving increasing weight
to workers with higher incomes and by applying the percentage ranges
provided by DOL officials. 

      SOURCES OF UNCERTAINTY
------------------------------------------------------- Appendix I:1.3

There are two sources of uncertainty in our estimation of the
potentially exempt population. 

The first source of uncertainty is methodological and related to the
probability of exemption for each occupation.  DOL officials provided
us with one of four ranges of likelihood of exemption for each
occupation.  Our low estimate is based on the lowest likelihood of
exemption in each of the four ranges and our high estimate is based
on the highest likelihood in each range. 

The second source of uncertainty is sampling error, due to our use of
the CPS survey of households.  Rather than counting the number of
employees actually classified as exempt by employers, we estimated
how many employees are likely to be classified as exempt, based on
the occupational classifications and income reported in the CPS
sample. 

The high and low estimates we report reflect the methodological
uncertainty, which is much larger than the sampling error.  For
example, consider table 2.  The table shows that the high estimate of
workers covered by the white-collar exemptions in 1983 is 17 million. 
For the same year, our low estimate is 12 million workers.  The
difference between the high and low estimate--5 million workers--is
the amount of uncertainty due to methodological process we describe
as our first source of uncertainty.  On the other hand, the sampling
error--the second source of uncertainty--involves an estimated error
of plus or minus 700,000 people. 

   IDENTIFICATION OF STATUTORY AND
   REGULATORY CHANGES
--------------------------------------------------------- Appendix I:2

We reviewed legislative references to identify changes to the FLSA. 
For the regulatory changes, we reviewed annual publications of the
Code of Federal Regulations for the period 1959 to the present; for
the period 1938 to 1959, we relied on three DOL hearing reports
provided to us by DOL. 

   DETERMINATION OF THE MAJOR
   CONCERNS OF EMPLOYERS AND
   EMPLOYEES
--------------------------------------------------------- Appendix I:3

To obtain information on the major concerns of employers and
employees about the white-collar exemptions, we used four sources: 
(1) review of federal court decisions involving white-collar
exemption issues over the 5-year period from January 1994 through
December 1998; (2) review of DOL compliance cases related to
white-collar exemptions that were closed in the 2-year period between
January 1997 through December 1998 in four DOL field offices; (3)
interviews with individuals and groups representing the interests of
employers; and (4) interviews with individuals and groups with
knowledge about the interests of employees.  In the paragraphs below,
we present more detail on each of these sources. 

      REVIEW OF FEDERAL COURT
      CASES
------------------------------------------------------- Appendix I:3.1

We reviewed 166 judicial opinions from federal appellate and district
courts involving litigation related to the white-collar exemptions
during the 5-year period from 1994 through 1998.  To identify these
cases, we conducted a computer search of federal opinions in this
period.\30 We categorized each of the 166 cases as being related to
one of four issues:  the salary-basis test, the executive
classification, the administrative classification, or the
professional classification.  For cases involving more than one
white-collar exemption issue, we selected one issue to be the primary
issue.  In addition to the primary issue, we also collected data on
other circumstances of each case, such as who prevailed on the issue
and whether the case involved multiple employees suing their
employer. 

--------------------
\30 To avoid double counting, we eliminated district court cases that
were later heard by an appellate court, and we only used the
appellate decision in our sample. 

      REVIEW OF DOL COMPLIANCE
      CASES
------------------------------------------------------- Appendix I:3.2

We supplemented our review of federal court decisions with a review
of 66 DOL compliance cases from 4 of its 57 District Offices.  At our
request, the four District Offices--in the Northeast (Boston), South
(Richmond), Midwest (Chicago), and West (Los Angeles)--identified
compliance cases closed in the 2-year period between January 1997 and
December 1998 that involved white-collar issues.  Each office
identified cases that we reviewed and discussed with district
managers and investigators who were familiar with the cases.  As with
the federal court decisions, we categorized each of the cases as
being related to one of four issues:  the salary-basis test, the
executive classification, the administrative classification, or the
professional classification. 

      EMPLOYER PERSPECTIVES
------------------------------------------------------- Appendix I:3.3

To identify the key concerns of employers related to the white-collar
exemptions, we met with private and public employers, trade
associations, DOL officials, and various legal and economics experts. 
The Society for Human Resource Management, a group of human resource
managers from companies with over 100 employees, organized two group
sessions--one with banking and insurance human resource managers, and
the other with manufacturing human resource managers.  We also met
with the Labor Policy Association--a group representing companies
with business operations in the United States and with more than
2,500 employees each.  In addition to these company managers, we met
with attorneys and associations representing manufacturing companies,
retail companies, and small businesses.  On the public employment
side, we met with representatives from various cities and public
employer groups, as well as separately with four legal experts
representing public employers across the country.  Finally, we met
with DOL officials in the Wage and Hour Division and the Office of
the Solicitor. 

      EMPLOYEE PERSPECTIVES
------------------------------------------------------- Appendix I:3.4

To identify the concerns of employees, we met with union
representatives and officials, DOL officials, and legal and economic
experts.  We met with the American Federation of Labor-Congress of
Industrial Organizations and officials from nine different unions and
trade union representatives.  We also interviewed several legal and
economic experts who specialize in issues relating to the FLSA and,
in particular, white-collar exemptions. 

STATUTORY AND REGULATORY HISTORY
OF FLSA WHITE-COLLAR EXEMPTIONS
========================================================== Appendix II

The Fair Labor Standards Act at sec.  13(a)(1) established the
so-called white-collar exemptions:  any employee in a "bona fide
executive, administrative, or professional capacity" is exempted from
the minimum wage and overtime requirements of the FLSA.  After the
original enactment of the FLSA in 1938, the Department promulgated
regulations (at 24 C.F.R.  part 541) defining the terms included in
section 13(a)(1).  Since 1938, there have been changes to both the
statute and the regulations.  In the following tables, we outline
these changes. 

                                        Table II.1
                         
                            Statutory History of FLSA Section
                           13(a)(1) on White-Collar Exemptions,
                                        1938-1998

Year   Legislation   Summary of change
-----  ------------  --------------------------------------------------------------------
1938   Fair Labor    As originally enacted in 1938, sec. 13(a)(1) provided that the
       Standards     minimum wage (section 6) and overtime provisions (section 7) of the
       Act of 1938   FLSA did not apply to "any employee employed in a bona fide
       (P.L. 75-     executive, administrative, professional, or local retailing
       718)          capacity."

1961   Fair Labor    This amendment limited the exemption for retail employees. It
       Standards     eliminated the separate exemption category for workers employed in a
       Amendments    "local retailing capacity." This separate exemption was replaced
       of 1961       with a proviso that an employee of a retail or service establishment
       (P.L. 87-     could not be excluded from the definition of executive or
       30)           administrative capacity because of the amount of time devoted to
                     other than executive or administrative activities if less than 40
                     percent of his or her work hours included such activities.

1966   Fair Labor    In 1966, public educational institutions were made subject to the
       Standards     requirements of the FLSA. However, this amendment exempted academic
       Amendments    administrative personnel and teachers in elementary and secondary
       of 1966       schools from these requirements by expressly including them under
       (P.L. 89-     sec. 13(a)(1).
       601)

1972   Education     This amendment made the equal-pay provision of the FLSA at sec. 6(d)
       Amendments    applicable to employees who were otherwise exempt from the FLSA
       of 1972       under sec. 13(a)(1).
       (P.L. 92-
       318)

1996   Small         This amendment, enacted into law at sec. 13(a)(17) of the FLSA,
       Business Job  contians the description of computer professionals who are exempt
       Protection    employees. By its terms, employees who meet the duties test of the
       Act of 1996   exemption are exempt from the FLSA as long as they earn not less
       (P.L. 104-    than $27.63 per hour.
       188)
-----------------------------------------------------------------------------------------
Source:  GAO analysis of applicable laws and legislative history. 

                                        Table II.2
                         
                          Regulatory History of 29 CFR Part 541,
                          Defining FLSA White-Collar Exemptions,
                                        1938-1998

      Federal Register
      notice of         Part 541
      regulation        subsections
Year  change            affected          Summary of change
----  ----------------  ----------------  -----------------------------------------------
1938  3 Fed. Reg.       Regulation        The Fair Labor Standards Act of 1938 in sec.
      2518,             included subsec.  13(a)(1) exempted "any employee employed in a
      Oct. 20, 1938     1-5               bona fide executive, administrative,
                                          professional or local retailing capacity" from
                                          the requirements of the Act. To define these
                                          terms, part 541 was added to the Department
                                          regulations in 29 C.F.R. chapter V. The new
                                          regulations defined these exempt employees, as
                                          follows:
                                          --The new regulation combined executive and
                                          administrative at sec. 541.1 to include any
                                          employee whose primary duty is the "management
                                          of the establishment, or a customarily
                                          recognized department thereof" and who
                                          "customarily and regularly directs the work of
                                          other employees therein." It also set a salary
                                          level--to be exempted, an employee had to be
                                          compensated at not less than $30 per week.
                                          --Separate definitions were included for
                                          professional employees (sec. 541.2:
                                          "customarily and regularly engaged in work
                                          predominantly intellectual and varied in
                                          character as opposed to routine mental, manual,
                                          mechanical or physical work") and retail
                                          employees (sec. 541.3: "customarily and
                                          regularly engaged in making retail sales").
                                          There was no salary limit for either
                                          professional or retail employees.

1940  5 Fed. Reg.       Regulation        In accommodating the views of interested
      4077,             revised subsec.   employers, employees, trade associations, and
      Oct. 15, 1940     1-5, adding new   unions, the Department revised the definitions
                        sec. 2            of executive, administrative, and professional
                                          to include separate duties and salary tests for
                                          each type of exempt employee. These revisions
                                          included the following:
                                          --For executive employees, the revised
                                          regulations at sec. 541.1 retained much of the
                                          original language, but specified that the
                                          employee could not spend more than 20 percent
                                          of the work-week on "work of the same nature as
                                          performed by nonexempt employees" unless in
                                          "sole charge" of an independent establishment.
                                          The executive was to be paid on a salary basis
                                          at least $30 per week.
                                          --Administrative employees were newly defined
                                          in sec. 541.2 as those employees who performed
                                          nonmanual work that required the exercise of
                                          discretion and independent judgment. The
                                          administrative employee was to be paid on a
                                          salary or fee basis at least $200 per month.
                                          --For professional employees, the regulations
                                          in sec. 541.3 added a new requirement that they
                                          either had knowledge of an advanced type in a
                                          field of science or learning or performed work
                                          predominantly original and creative in
                                          character. They also had to be paid on a salary
                                          or fee basis at least $200 month unless they
                                          were licensed doctors or lawyers.

1942  7 Fed. Reg.       Revised subsec.   This revision added para. (b)(4) to the
      332,              2                 definition of administrative employees,
      Jan. 17, 1942                       specifically exempting employees engaged in
                                          transporting goods or passengers for hire.
                                          According to a 1949 DOL report, this provision
                                          was intended to exempt employees engaged in
                                          ferrying airplanes. (This provision was
                                          eliminated in 1949.)

1949  14 Fed. Reg.      Revised subsec.   After 10 years of administrative experience,
      7705,             1-6               the Department once again revised the
      Dec. 24, 1949                       regulations. For the most part, these changes
                                          resulted from recommendations from the DOL's
                                          regional directors and field staff. Among other
                                          things, the changes included the following:
                                          --A special "upset" test was added to the tests
                                          determining administrative, executive, or
                                          professional employees: if the employee made at
                                          least $100 per week (a "high-salaried"
                                          employee), then he or she only had to meet a
                                          "short-cut" version of the original long duties
                                          test.
                                          --The basic salary test was increased:
                                          executives had to make at least $55 per week;
                                          administrators, at least $75 per week; and
                                          professionals, at least $75 per week.
                                          --For executives, the regulations specified
                                          that they must supervise two or more employees.

1949  14 Fed. Reg.      Added new         To respond to requests from its field staff for
      7730,             Subpart B--       explanatory material on the 541 regulations,
      Dec. 28, 1949     Interpretations   the DOL published an explanatory bulletin. This
                                          bulletin, added to the regulation as Subpart B,
                                          contained statements of general policy directly
                                          related to the 541 regulations. Among other
                                          things, the bulletin:
                                          --Elaborated upon the terms included in subsec.
                                          1-6; such terms included primary duty
                                          (541.103), sole charge (541.113) and salary
                                          basis (541.118).
                                          --Added new terms to the regulations, such as
                                          working foremen (541.115) and combination
                                          exemptions (541.600).
                                          --Applied the regulations to specific
                                          professions, such as newspaper reporters
                                          (541.303f) and radio announcers (541.303e).

1953  18 Fed. Reg.      Added subsec.     Subsec. 5a made the requirement that exempt
      3930,             5(a) and 601      employees be paid on a salary basis not
      July 7, 1953;                       applicable to employees in the motion-picture-
      error in                            producing industry who were compensated at
      original notice                     certain minimum rate, and subsec. 601 explained
      corrected by 18                     how to apply this provision.
      Fed. Reg. 4098,
      July 14, 1953

1954  19 Fed. Reg.      Revised subsec.   Subsec. 118, which explains the salary-basis
      4405,             118               test, was revised, in principal part, as
      July 17, 1954                       follows:
                                          --Subsec. 118(a) specified that an employee
                                          must receive a full salary "without regard to
                                          the number of days or hours worked." This
                                          revision explained the application of this rule
                                          to different pay deductions; for example, an
                                          employer cannot deduct pay for a lack of work
                                          (541.118(a)(1)) or for personal absences of
                                          less than 1 day (541.118(a)(2)) or for jury
                                          duty (541.118(a)(4)).
                                          --In addition, a new subsec. 118(a)(6) allowed
                                          an employer, under certain circumstances, to
                                          correct improper pay deductions if the employer
                                          reimbursed the employee and promised to comply
                                          in the future (the so-called window of
                                          correction).

1958  23 Fed. Reg.      Revised subsec.   This revision increased the salary tests for
      8962, Nov. 18,    1, 2, and 3       exempt employees:
      1958                                --Executive employees had to be paid at least
                                          $80 per week; administrative employees, at
                                          least $95 per week; and professional employees,
                                          at least $95 per week.
                                          --The so-called "upset test" for high-salaried
                                          employees was also increased to $125 per week
                                          for all three types of exempt workers.

1959  24 Fed. Reg.      Revised subsec.   This revision adjusted the provisions in
      581,              100, 117, 118,    Subpart B--Interpretations to match the salary
      Jan. 27, 1959     119, 200, 211,    test increases published in November 1958.
                        300, 311, 313,
                        315, 600

1961  26 Fed. Reg.      Revised subsec.   This revision implemented the Fair Labor
      8635,             1, 2, 3, 99,      Standards Amendments of 1961, which eliminated
      Sept. 15, 1961    100, 101, 105,    the exemption for employees in a "local
                        109, 112, 113,    retailing capacity." The amendments replaced it
                        114, 200, 209,    with a proviso that retail employees could not
                        300, 308;         be excluded from the executive or
                        revoked subsec.   administrative exemptions because of the amount
                        4, 400, 401,      of time they spent performing activities that
                        402, 403          are not managerial or administrative if less
                                          than 40 percent of the employee's time was
                                          devoted to such activities.

1963  28 Fed. Reg.      Revised subsec.   This revision updated Subpart B--
      9505, Aug. 30,    included 1, 2,    Interpretations to include illustrative
      1963,             3, 100, 105,      examples relating to retail work. It also
      typographical     108, 109, 112,    increased the salary test for exempt
      error in          113, 117, 118,    employees:
      original notice   119, 200, 201,    --Executive employees had to be paid a salary
      corrected by 28   202, 205, 207,    of at least $100 per week; administrative
      Fed. Reg. 14423,  209, 211, 300,    employees, at least $100 per week; and
      Dec. 28, 1963     311, 315, 600;    professional employees, at least $115 per
                        added subsec. 5b  week.
                                          --The upset test triggering the short duties
                                          test was increased to $150 per week for all
                                          three types of exempt employees.
                                          --The increased salary test was not effective
                                          for retail employees until September 3, 1965.

1967  32 Fed. Reg.      Revised subsec.   This revision was made to implement the changes
      7823,             included 1, 2,    in the law made by the Fair Labor Standards
      May 30, 1967      3, 112, 200,      Amendments of 1966, which made public
                        201, 202, 300,    educational institutions subject to the FLSA
                        302, 304, 307,    for the first time. It also amended the Act to
                        314, 315, 602     specifically include academic administrative
                                          personnel and teachers in elementary and
                                          secondary schools in sec. 13(a)(1). The
                                          regulatory changes pursuant to this amendment
                                          included, among other things:
                                          --Subsec. 2 was expanded to specify an
                                          administrative exemption for performance of
                                          administrative functions in a school system.
                                          --Subsec. 3 included a professional exemption
                                          for teachers in a school system or educational
                                          establishment.
                                          --Teachers, as well as doctors and lawyers,
                                          were excepted from the requirement for a
                                          minimum salary level.

1970  35 Fed. Reg.      Revised subsec.   These revisions increased the salary test for
      883,              1, 2, 3, 5b,      exempt employees:
      Jan. 22, 1970     100, 117, 118,    --Executive employees had to be paid at least
      (35 Fed. Reg.     119, 200, 211,    $125 per week; administrative employees, at
      3220,             214, 300, 311,    least $125 per week; and professional
      Feb. 20, 1970,    313, 315, 600;    employees, at least $140 per week.
      extended the      revoked subsec.   --The upset test triggering the short duties
      effective date    200 and 300       test was increased for all three types of
      of the salary                       employees to $200 per week.
      increase)

1971  36 Fed. Reg.      Revised subsec.   These revisions to Subpart B included
      22976,            included 103,     additional explanatory guidelines for
      Dec. 2, 1971      207, 302          paramedical and data-processing employees, as
                                          well as for the professional exemption.

1973  38 Fed. Reg.      Revised subsec.   These revisions reflect the changes made in the
      11390,            included 1, 2,    law by the Education Amendments of 1972, which
      May 7, 1973       3, 52, 117, 118,  made the equal-pay provision of the FLSA
                        119, 214, 311,    applicable to otherwise-exempt employees.
                        313, 315, 601

1975  40 Fed. Reg.      Revised subsec.   These revisions increased the salary test for
      7092,             1, 2, 3, 117,     exempt employees:
      Feb. 19, 1975     118, 119, 211,    --Executive employees had to be paid at least
                        214, 311, 313,    $155 per week; administrative employees, at
                        315, 601          least $155 per week; and professional
                                          employees, at least $170 per week.
                                          --The upset test triggering the short duties
                                          test was increased for all three types of
                                          employment to $250 per week.

1981  46 Fed. Reg.      Revised subsec.   These revisions, which were indefinitely
      3010,             1, 2, 3, 117,     postponed by Presidential memorandum of Jan.
      Jan. 13, 1981     118, 119, 211,    29, 1981, would have increased the salary test
      (postponed        214, 311, 313,    for exempt employees:
      indefinitely);    315, 601;         --Executive employees would have had to be paid
      46 Fed. Reg.      revoked subsec.   at least $225 per week; administrative
      11972,            52                employees, at least $225 per week; and
      Feb. 12, 1981;                      professional employees, at least $250 per week.
      and 46 Fed. Reg.                    In 2 years, these base salaries would have been
      11972, Feb. 12,                     raised to $250 per week for executive and
      1981                                administrative employees, and to $280 per week
                                          for professional employees.
                                          --The upset test triggering the short duties
                                          test would have increased for all three types
                                          of employment to $320 per week. In 2 years,
                                          this test would have been raised to $345 per
                                          week.

1991  56 Fed. Reg.      Added subsec. 5d  This new subsection was added in response to
,     45824,                              conflicting court decisions and the
1992  Sept. 6, 1991,                      accompanying confusion, and the need to treat
      as revised and                      public employees differently than private
      finalized at 57                     employees because of the requirement for public
      Fed. Reg. 37666,                    accountability. It made the requirement of
      Aug. 19, 1992                       payment on a salary basis generally, but not
                                          completely, inapplicable to exempt state and
                                          local employees.

1992  56 Fed. Reg.      Revised subsec.   This revision implemented sec. 2 of P.L. 101-
      8250, Feb. 27,    3, 5c, 302, 312;  583 which directed the Secretary of Labor to
      1991, as          added subsec.     promulgate regulations that permitted computer
      finalized at 57   303               systems analysts, computer programmers,
      Fed. Reg. 46742,                    software engineers, and other similarly skilled
      Oct. 9, 1992                        workers to be classified as exempt employees.
      (corrected by 57                    As specified by the law, these revisions
      Fed. Reg.                           exempted workers whose duties met the new test
      47163,                              set forth in the regulation and, if they were
      Oct. 14, 1992)                      paid on a salary basis, their pay met the
                                          specified salary test levels; or, if they were
                                          paid on an hourly basis, their regular rate of
                                          pay exceeded 6-1/2 times the minimum wage.\a
-----------------------------------------------------------------------------------------
Source:  GAO analysis of applicable regulations. 

\a P.L.  104-188 subsequently set the hourly wage that would allow
employers to claim an exemption for computer workers at $27.63 per
hour, but no corresponding change was made in the regulations. 

(See figure in printed edition.)Appendix III
COMMENTS FROM THE DEPARTMENT OF
LABOR
========================================================== Appendix II

(See figure in printed edition.)

(See figure in printed edition.)

(See figure in printed edition.)

GAO CONTACTS AND STAFF
ACKNOWLEDGMENTS
========================================================== Appendix IV

GAO CONTACTS

Larry Horinko, (202) 512-7001
Nancy Peters, (202) 512-9065

STAFF ACKNOWLEDGMENTS

In addition to those mentioned above, Carol Patey, Bill Hansbury,
Rich Kelley, Kelly Mikelson, Charlie Jeszeck, and Robert Crystal made
key contributions to this report. 

*** End of document. ***