Employment-Based Health Insurance: Medium and Large Employers Can
Purchase Coverage, but Some Workers Are Not Eligible (Letter Report,
07/27/98, GAO/HEHS-98-184).

Pursuant to a legislative requirement, GAO reviewed the extent to which
classes of large employers in different states have access to health
insurance and the circumstances surrounding the lack of access, if any,
to coverage, focusing on: (1) the extent to which medium and large
employers in different categories and states have access to health
insurance and the barriers, if any, these employers face in seeking
health insurance; (2) major factors that affect employers' decisions to
offer health insurance; and (3) the extent to which employees are
eligible for their employer-provided health plans.

GAO noted that: (1) virtually all medium and large employers have access
to group health insurance, and about 90 percent actually offer health
coverage to their employees; (2) the larger the firm the more likely it
is to offer health insurance; (3) employers that do not offer health
insurance are likely to be influenced by a variety of factors such as
firm size, the wage level, and health insurance premiums; (4)
sponsorship of health insurance by medium and large employers varies
moderately by industry sector and somewhat more widely by state; (5)
across the states, estimated offer rates among medium and large
employers ranged from 72 percent in Wyoming to close to 100 percent in
14 states in 1993; (6) the highest offer rates were concentrated in the
Northeast, including the six New England states, New Jersey, and
Pennsylvania; (7) the propensity of an employer to offer health
insurance depends not only on the firm's size but on other, often
interrelated workforce and firm characteristics, including employee
earnings, labor turnover, and health insurance cost; (8) by contrast,
some medium employers may be less likely to offer coverage if they
believe that lower-wage employees would not enroll in the plans; (9) in
addition, medium employers may be less likely to offer coverage if
insurers charge high premiums to cover the higher administrative costs
of handling smaller groups or the greater variability of a smaller
group's costs due to a catastrophic medical case; (10) data on the
prevalence of these circumstances are not available; (11) some employees
of medium and large firms do not have access to their employers'
sponsored health plan because of eligibility requirements; (12) the
eligibility rate of part-time employees was about 31 percent in contrast
to a rate of more than 80 percent for full-time workers in 1997; (13)
although the majority of part-time employees nationwide have
coverage--mostly through another family member's job or nonemployment
sources--5.4 million were uninsured in 1997; and (14) uninsured
part-time employees represented about 27 percent of the roughly 19.9
million uninsured wage and salary workers that year.

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

 REPORTNUM:  HEHS-98-184
     TITLE:  Employment-Based Health Insurance: Medium and Large 
             Employers Can Purchase Coverage, but Some Workers Are Not
             Eligible
      DATE:  07/27/98
   SUBJECT:  Employee medical benefits
             Health insurance
             Health insurance cost control
             Insurance premiums
             Eligibility criteria
IDENTIFIER:  National Center for Health Statistics National Employer 
             Health Insurance 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 Congressional Committees

July 1998

EMPLOYMENT-BASED HEALTH INSURANCE
- MEDIUM AND LARGE EMPLOYERS CAN
PURCHASE COVERAGE, BUT SOME
WORKERS ARE NOT ELIGIBLE

GAO/HEHS-98-184

Employment-Based Health Insurance

(101591)


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

  AHCPR - Agency for Health Care Policy and Research
  BLS - Bureau of Labor Statistics
  HHS - Department of Health and Human Services
  HIPAA - Health Insurance Portability and Accountability Act of 1996
  MEPS - Medical Expenditure Panel Survey
  NCHS - National Center for Health Statistics
  NEHIS - National Employer Health Insurance Survey

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


B-278073

July 27, 1998

Congressional Committees

Most Americans younger than 65 have health insurance, largely through
their own or a family member's employment, but about 41 million
people lacked health insurance in 1996.  Concerned about insurers'
willingness to offer policies to small employers and certain
individuals, the Congress passed the Health Insurance Portability and
Accountability Act of 1996 (HIPAA) (P.L.  104-191, Aug.  21, 1996) to
improve access to health insurance.  HIPAA contains provisions
related to initial coverage, renewability, and portability of private
health insurance.  These provisions vary in their application to
large and small employers who buy or wish to buy coverage for their
employees and individuals attempting to purchase health insurance on
their own.  For example, HIPAA requires health insurance carriers
operating in the "small group" market to accept any small employer
that applies for coverage, regardless of the group's health status. 
This guaranteed-access provision does not extend to medium and large
firms--those employing more than 50 employees.  In addition, HIPAA
does not directly address premium rates. 

To help the Congress gain a better understanding of access issues in
the larger group market, HIPAA required that we report on the extent
to which classes of large employers in the different states have
access to health insurance and the circumstances surrounding the lack
of access (if any) to coverage.\1 This report provides the results of
our study.\2 As agreed with your offices, we examined (1) the extent
to which medium and large employers (those with 51 or more employees)
in different categories and states have access to health insurance
and the barriers (if any) these employers face in seeking health
insurance, (2) major factors that affect employers' decisions to
offer health insurance, and (3) the extent to which employees are
eligible for their employer-provided health plans.\3

To address these objectives, we analyzed government and private
sector survey data on (1) the number of employers that offer health
insurance to their employees by firm size, industry sector, and
location and (2) employee eligibility rates by employment status,
such as full-time versus part-time schedules and permanent and
temporary work arrangements.  These data pre-dated the implementation
of HIPAA's provisions concerning access, renewability, and
portability of private health insurance.  We used the percentage of
employers that offer health insurance--employers' offer rate--as a
proxy for the extent of employers' access to the health insurance
market.  The offer rate is a widely used indicator of employers'
access to health insurance but it is incomplete in that it measures
only "realized" access.  Some employers that do not currently offer
health coverage nonetheless would be able to obtain coverage if they
chose to purchase it; they have potential access that is not
realized.  In contrast, other employers that do not offer coverage to
their employees may in fact lack access to health insurance. 
Existing government and private sector surveys of employee health
benefits do not contain information that would permit employers with
potential but unrealized access to be distinguished from those
lacking access.  To fill this gap, we interviewed industry experts
and representatives from insurance agencies and brokers, state
regulatory agencies, and employers in different states.  Information
from these interviews also supplemented our literature review on
factors influencing employers' decisions to provide employer-based
health insurance coverage.  (See app.  I for details on our
methodology.) We conducted our work between September 1997 and June
1998 in accordance with generally accepted government auditing
standards. 


--------------------
\1 HIPAA also requires that the Secretary of Health and Human
Services (HHS) report to the Congress on the extent to which large
employers seeking to obtain health coverage in different states are
able to do so.  HHS' report to the Congress, due every 3 years, is to
be based, among other things, on information that HHS will request
each state to submit no later than December 31, 2000. 

\2 Addressees are listed at the end of this letter. 

\3 HIPAA defines large employers as those with an average of at least
51 employees during the preceding calendar year.  We use the terms
medium and large employers to emphasize the broad range of employers
included in HIPAA's definition of large employer. 


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

Virtually all medium and large employers have access to group health
insurance, and about 90 percent actually offer health coverage to
their employees.  The larger the firm the more likely it is to offer
health insurance.  For example, 1996 survey data show that about 96
percent of firms with 100 to 199 employees offered health coverage
compared with about 91 percent of firms with 50 to 99 employees. 
Employers that do not offer health insurance are likely to be
influenced by a variety of factors such as firm size, the wage level,
and health insurance premiums. 

Sponsorship of health insurance by medium and large employers varies
moderately by industry sector and somewhat more widely by state.  For
example, health insurance offer rates of these employers ranged from
88 percent in the agriculture, forestry, and fishing industry sector
to 99 percent in wholesale trade in 1993--the most recent year for
which data are available by firm size for different industries and
states.  Across the states, estimated offer rates among medium and
large employers ranged from 72 percent in Wyoming to close to 100
percent in 14 states in 1993.  The highest offer rates were
concentrated in the Northeast, including the six New England states,
New Jersey, and Pennsylvania. 

The propensity of an employer to offer health insurance depends not
only on the firm's size but also on other, often interrelated
workforce and firm characteristics, including employee earnings,
labor turnover, and health insurance cost.  For example, the
likelihood of health benefits being offered increases with a higher
proportion of high-wage employees in the workforce and a lower
turnover rate.  By contrast, some medium employers may be less likely
to offer coverage if they believe that lower-wage employees would not
enroll in the plans.  In addition, medium employers may be less
likely to offer coverage if insurers charge high premiums to cover
the higher administrative costs of handling smaller groups or the
greater variability of a smaller group's costs due to a catastrophic
medical case.  Data on the prevalence of these circumstances are not
available. 

Some employees of medium and large firms do not have access to their
employers' sponsored health plans because of eligibility
requirements.  For example, many part-time and temporary employees
are not eligible for coverage under their employers' health plans. 
The eligibility rate of part-time employees was about 31 percent in
contrast to a rate of more than 80 percent for full-time workers in
1997.  Although the majority of part-time employees nationwide have
coverage--mostly through another family member's job or nonemployment
sources--5.4 million were uninsured in 1997.  Uninsured part-time
employees represented about 27 percent of the roughly 19.9 million
uninsured wage and salary workers that year. 


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

Employment-based health insurance is the leading source of private
health insurance in the United States.  In 1996, about 64 percent of
the population under 65 years old had employment-based health
insurance.  This employer-based system of health insurance is almost
entirely voluntary.\4

The majority of private business establishments in the United States
are small, but most workers are employed by medium and large
employers.  A business establishment is a single location where
economic activity is conducted.  As shown in figure 1, establishments
with fewer than 20 workers constituted about 88 percent of the total
but accounted for only about 26 percent of the U.S.  employment in
1996.  In contrast, medium and large establishments, with 50 or more
employees, represented about 5 percent of the total but accounted for
about 58 percent of U.S.  employment.  Figure 1 also shows that more
than one-quarter of all employees work for establishments employing
250 or more people, but these businesses represent less than 1
percent of all establishments.  Three industries--services, retail
trade, and manufacturing--accounted for the majority (about 73
percent) of all private sector employment in 1996.  (See fig.  2.)

   Figure 1:  Relationship Between
   Private Sector Establishment
   Size and Employment
   Concentration, First Quarter
   1996

   (See figure in printed
   edition.)

Source:  U.S.  Department of Labor, Bureau of Labor Statistics,
published November 1997. 

   Figure 2:  Percentage of
   Private Sector Employment by
   Industry, 1996

   (See figure in printed
   edition.)

Source:  U.S.  Department of Labor, Bureau of Labor Statistics,
published November 1997. 


--------------------
\4 Hawaii is the only state permitted under the Employee Retirement
Income Security Act of 1974 to mandate that employers offer health
insurance. 


   MEDIUM AND LARGE EMPLOYERS HAVE
   NEARLY UNIVERSAL ACCESS TO
   HEALTH INSURANCE
------------------------------------------------------------ Letter :3

Survey data that we analyzed and our interviews with brokers
indicated that virtually all medium and large firms have access to
health insurance coverage.  Their estimated offer rates of more than
90 percent show that these firms typically take advantage of their
access by offering coverage.  While the estimated health insurance
offer rates between large and small employers differ markedly, the
offer rates among medium and large employers vary moderately by firm
size and industry and somewhat more widely by state. 

As shown in table 1, the estimated percentage of medium and large
employers that offered health insurance in 1996 ranged from 91 to 99
percent, with larger firms more likely to offer health insurance than
medium firms.  In contrast, the estimated offer rate among employers
with fewer than 50 employees averaged less than 50 percent, according
to a recent KPMG Peat Marwick health benefits survey.\5 Medium
employers that do not offer health coverage generally have access to
health insurance but may decide for various reasons that they do not
need to offer health benefits as part of their employee compensation
package or may consider the cost of providing health coverage to be
too high relative to their profits or employees' wages. 



                          Table 1
          
            Estimated Percentage of Private and
          Public Sector Employers Offering Health
                Insurance by Firm Size, 1996

                                             Percentage of
                                            firms offering
Number of employees in firm               health insurance
--------------------------------------  ------------------
Less than 50                                            48
50-99                                                   91
100-199                                                 96
200 and more                                            99
All                                                     51
----------------------------------------------------------
Source:  KPMG Peat Marwick. 

The most recent data available that permit analysis of health
insurance offer rates by industry sector and states across firm sizes
come from the 1993 National Employer Health Insurance Survey (NEHIS),
managed by HHS' National Center for Health Statistics (NCHS).  These
data show modest variation in the percentage of employers offering
health insurance across industry sectors.  Among medium and large
employers in different industries, the estimated offer rates in 1993
ranged from 88 to 99 percent.  The industries with the highest
estimated offer rates (more than 95 percent) were wholesale trade;
mining; manufacturing; finance, insurance, and real estate; and
transportation, communications, and utilities.  As shown in figure 3,
slightly lower offer rates (between 88 and 94 percent) prevailed in
the retail trade; construction; services; and agriculture, forestry,
and fishing industries.  The industries with estimated lower offer
rates are characterized by higher rates of labor turnover and
prevalence of temporary and part-time workers.  A similar
relationship between offer rates and type of industry was found among
small employers, but small employers' offer rates were markedly lower
across all industries, ranging from 29 to 56 percent. 

   Figure 3:  Estimated Percentage
   of Establishments Offering
   Health Insurance by Type of
   Industry and Firm Size, 1993

   (See figure in printed
   edition.)

Source:  NEHIS. 

Variation by state is wider than by industry.  At the state level,
the estimated offer rate among medium and large employers ranged from
a low of about 72 percent in Wyoming and Texas to almost 100 percent
in 14 states.  The higher offer rates were concentrated in the
Northeast, including the six New England states, New Jersey, and
Pennsylvania.  Because these estimates are based on survey data, the
estimate derived from the sample may be higher or lower than the true
percentage of all employers in the state that offer health insurance. 
Because of the small number of establishments surveyed in certain
states, the degree of imprecision for these estimates, as measured by
their confidence intervals, can be quite large.  Details on the
estimated offer rates by firm size in each state and associated
confidence intervals for each firm size grouping are in appendix II. 

Survey data on offer rates presented above estimate the extent of
realized access to insurance, but employers that do not offer health
coverage often have potential access.  That is, such employers have a
choice.  According to insurers, insurance brokers, and employers that
we interviewed, employers with more than 50 workers generally can get
an insurer to write a policy covering their workforce.\6 However,
insurers and brokers we interviewed pointed out that the price of
such coverage for employers that have close to 50 employees might
differ substantially--from employer to employer and from insurer to
insurer. 

Survey data, studies, and interviews with market participants
indicate that variations in health insurance offer rates by firm
size, industry, and state reflect decisions that medium and large
employers make about offering health coverage, not difficulties they
face in obtaining access to health insurance.  As discussed below,
multiple factors, including the characteristics of a firm and its
workforce and the cost of health insurance, influence employers'
health insurance decisions. 


--------------------
\5 KPMG Peat Marwick also reported the following offer rates among
small firms in 1996:  1 to 9 employees, 42 percent; 10 to 24
employees, 78 percent; and 25 to 49 employees, 90 percent. 

\6 Other evidence on the extent of access is the number of employers
that are offered health insurance.  According to an analysis by
Professor Gail A.  Jensen (from Wayne State University) of a 1993
national survey, over 80 percent of employers with 50 to 199
employees reported receiving at least one solicitation to buy health
insurance in the last 6 months. 


   DIFFERENCES IN OFFER RATES
   REFLECT FIRMS' WAGE LEVELS AND
   OTHER TRAITS
------------------------------------------------------------ Letter :4

Although the likelihood of a firm's offering health insurance is
strongly linked with its size, whether an employer offers health
coverage to its workers also depends on the characteristics of its
workforce, the cost of providing health insurance, the firm's
financial condition, and the competitive environment in which it
operates.  Many of these factors are interrelated, and while research
studies we reviewed do not estimate the separate effect of these
factors, our analysis of survey data and research studies illustrates
the relationship between offer rates and certain characteristics of a
firm's workforce such as employee earnings and labor turnover.\7

Our analysis of 1993 NEHIS data supports other studies' findings that
employers' offer rates are lower when their workforces are
characterized by low wages and high labor turnover.  In addition, the
price of health coverage is likely to contribute to decisions that
some medium firms make regarding whether to offer health insurance. 


--------------------
\7 For further information on employer characteristics that have been
related to health insurance decisions, see NCHS, Employer-Sponsored
Health Insurance:  State and National Estimates, PHS 98-1005
(Hyattsville, Md.:  1997); and Len Nichols and others, Small
Employers:  Their Diversity and Health Insurance (Washington, D.C.: 
Urban Institute, 1997).  The labor turnover rate measures the
aggregate movement of persons into and out of jobs and between jobs. 


      THE ROLE OF WAGES
---------------------------------------------------------- Letter :4.1

NEHIS data show that employers with predominantly low-wage
employees--those earning less than $10,000 per year--are less likely
to offer health insurance to their workers.\8 The effect is generally
stronger the smaller the firm.  The difference in estimated health
insurance offer rates between employers whose workers typically
earned less than $10,000 and those whose workers typically earned
$10,000 or more a year in 1993 ranged from 10 percentage points among
employers with more than 200 employees to 33 percentage points among
employers with 51 to 100 employees.  (See table 2.)



                          Table 2
          
           Relationship Between Employers' Health
            Insurance Offer Rates and Employees'
                Earnings by Firm Size, 1993

                    (Numbers in percent)

                                 51 to    101 to      More
                    Up to 50       100       200  than 200
Annual wage level   employee  employee  employee  employee
of employees               s         s         s         s
------------------  --------  --------  --------  --------
Majority earned           19        61        73        87
 less than $10,000
Majority earned           51        94        95        97
 $10,000 or more
Difference between        32        33        22        10
 higher-and lower-
 wage firms
----------------------------------------------------------
Source:  NEHIS. 

The relatively low offer rate among employers with typically low-wage
employees is partly because the cost of health insurance represents a
substantial expense for both employers and employees.  According to
Mercer/Foster-Higgins' 1997 national survey of employers, the
nationwide average cost of providing health insurance to an employee
ranged from $3,200 per year for a health maintenance organization
plan to about $3,500 per year for an indemnity plan.  For the
employer, paying a significant portion of such premiums would add
considerably to the compensation of lower-wage workers or could lead
the employer to offset the cost of health insurance by reducing
wages.  For employees earning lower wages, paying the required share
of premium costs may be prohibitive. 

Contributions employers make toward employees' health benefits are
excluded from employees' taxable income.  However, this tax exclusion
is of lesser value to lower-wage employees who have a lower tax rate
than to high-wage employees.  Because low-wage workers do not benefit
as much as high-wage workers from the tax advantage associated with
employer-sponsored health insurance, many low-wage workers are likely
to prefer cash wages over health coverage.  As a result, lower-wage
workers may decline to enroll in an employer-based plan if one is
offered.  If few workers choose to participate in a plan, the firm
may decide not to offer coverage.\9


--------------------
\8 For further information on health insurance offer rates by
employee wage level and firm size, see NCHS, Employer-Sponsored
Health Insurance. 

\9 A factor contributing to such a decision is that carriers may
require that employers have a certain minimum percentage of eligible
employees to enroll in a group health plan.  This threshold
percentage typically varies between 75 and 100 percent. 


      THE ROLE OF LABOR TURNOVER
---------------------------------------------------------- Letter :4.2

Although NEHIS data do not measure the degree of labor turnover,
various studies have cited it as predictor of the likelihood that a
firm will offer health insurance.\10 Some research literature
indicates that firms with high labor turnover rates are less likely
to offer health insurance.  Quantitative analyses of the link between
labor turnover and firms' decisions to offer health insurance are
rare, but several health insurance industry experts we interviewed
indicated that this relationship exists.\11 Because labor turnover
varies by industry, this relationship can be illustrated by examining
industry data.  U.S.  Bureau of the Census data show that industries
that have labor turnover rates higher than the national average such
as agriculture, forestry and fishing; services; construction; and
retail trade had lower offer rates than industries with low turnover
rates.\12

This relationship is illustrated in figure 4, which shows that
industries with average monthly labor turnover rates around 10
percent or more had health insurance offer rates below 55 percent.\13
In contrast, industries with average monthly turnover rates below 7
percent had offer rates greater than 60 percent. 

   Figure 4:  Average Monthly
   Labor Turnover Rates by
   Industry Group and Estimated
   Employers' Offers of Health
   Insurance

   (See figure in printed
   edition.)

Source:  U.S.  Bureau of the Census and NEHIS data. 


--------------------
\10 See Paul B.  Ginsburg, Jon R.  Gabel, and Kelly A.  Hunt,
"Tracking Small-Firm Coverage, 1989-1996," Health Affairs, Jan.-Feb. 
1998, pp.  167-171, and Stephen H.  Long and M.  Susan Marquis, "Gaps
in Employment-Based Health Insurance:  Lack of Supply or Lack of
Demand?" in U.S.  Department of Labor, Health Benefits and the
Workforce (Washington, D.C.:  1992), pp.  37-42. 

\11 See Len M.  Nichols and others, Small Employers, p.  40 and Cori
E.  Uccello, Firms' Health Insurance Decisions:  The Relative Effects
of Firm Characteristics and State Insurance Regulations (Washington,
D.C.:  Urban Institute, July 1996). 

\12 U.S.  Bureau of the Census, Current Population Reports, Dynamics
of Economic Well-Being:  Labor Force, 1991-1993 (Washington, D.C.: 
Aug.  1995).  Labor turnover rates that exceed 10 percent are
considered high relative to the national average rate of 7 percent. 
The latest available data on labor turnover rates by industry sector
were collected by the U.S.  Bureau of the Census in 1991. 

\13 Because turnover data were not available by firm size, the
estimated offer rates cited in this section are for all firm sizes. 


      THE ROLE OF PREMIUM COSTS
---------------------------------------------------------- Letter :4.3

In health benefits surveys that examine reasons some employers
(especially those in smaller firms) do not offer health insurance,
employers have frequently cited high premium costs as a key factor in
their decisions.  However, data needed to determine the extent to
which costs affect the decisions medium and large employers make
regarding health coverage are lacking.\14 Although data are available
on premiums charged to employers who offer health coverage, data are
lacking on premium rates that firms not offering health insurance
would face if they sought to offer health coverage.  To address the
lack of data on actual premiums faced by non-offering firms,
researchers have developed various indirect methods but have applied
them only to data on small employers, the group with the lowest
percentage of employers offering health insurance.\15

Moreover, available data on average premiums may mask important
differences in the premiums charged to some medium firms that share
some of the characteristics of small employers.  For example, recent
data collected by a health benefits consulting firm show that health
insurance premiums do not vary greatly by employer size.  By
contrast, some older studies found that insurance companies charged
higher premiums to small companies because of the higher risks and
higher cost of marketing and claims administration involved in
providing coverage to small companies.\16 In addition, research based
on data from the 1980s showed that smaller employers have less power
to negotiate price discounts than larger employers.  However, some
studies that report recent data generally do not adjust for
differences in the depth and breadth of health benefits offered. 
Although smaller firms may be paying premiums similar to those paid
by larger firms, smaller firms (including those near the 50-employee
threshold) generally have less-generous health benefit packages with
higher deductibles.  In effect, these smaller firms are paying more
per dollar of benefits provided or claims reimbursed.\17

Some insurance brokers we interviewed said that some medium employers
(those closer to the 50-employee threshold) face greater variability
in health insurance claims cost and in premiums than larger
employers.  The extent of premium variation depends on how premiums
are set.  For example, a firm with 55 employees whose premiums are
based on experience rating--the group's historical medical costs--may
be charged higher per capita premiums in the future if an employee or
dependent were to incur catastrophic medical costs than a firm with
500 to 1,000 employees that also had a single very expensive case. 
In practice, the extent to which the premium increases for such a
firm depends, among other things, on whether the firm can negotiate
with its insurer for lower premium costs, the pressures of state
regulation that may limit premium variation or restrict premium
increases, and the rating practices of the insurer.\18


--------------------
\14 The role of costs in the small-group market has also become
increasingly difficult to assess because this market has been
changing rapidly.  Recent surveys indicate that the importance of
high premium costs as a factor in small firms' decisions to offer
coverage has been declining.  For a discussion of these trends, see
Paul B.  Ginsburg, Jon R.  Gabel, and Kelly A.  Hunt, "Tracking
Small-Firm Coverage, 1989-1996," Health Affairs, Jan.-Feb.  1998, and
Gail A.  Jensen and Michael A.  Morrisey, "Managed Care and the Small
Group Market," paper prepared for the conference:  Managed Care and
Changing Health Care Markets, American Enterprise Institute,
Washington, D.C., Apr.  10, 1997. 

\15 For example, a recent study of the role of premiums in the
decision to offer insurance uses an econometric approach to predict
these unobserved premiums.  The study obtained results for a sample
of Minnesota firms with less than 50 employees.  See Roger Feldman
and others, "The Effect of Premiums on the Small Firm's Decision to
Offer Health Insurance.", Journal of Human Resources, Vol.  32, No. 
4 (24) (Fall 1997), p.  635.  For information on other studies, see
Michael A.  Morrisey and Gail A.  Jensen, "State Small Group
Insurance Reform," pp.  71-90 in Robert Reich and William White,
Health Policy, Federalism, and the American States (Washington, D.C.: 
Urban Institute Press, 1996). 

\16 Some older studies are Stephen H.  Long and M.  Susan Marquis,
"Gaps in Employment-Based Health Insurance," and Arleen Leibowitz,
Cheryl Damberg, and Kathleen Eyre, "Multiple Employer Arrangements,"
in Health Benefits and the Workforce (Washington, D.C.:  U.S. 
Department of Labor, 1992). 

\17 For more information on average premiums for medium and small
firms, see Gail A.  Jensen and others, "The New Dominance of Managed
Care:  Insurance Trends in the 1990s," Health Affairs, Vol.16, No.  1
(Jan.-Feb.  1997), pp.  125-136. 

\18 For information on state approaches to premium rate restrictions
among small employers, see Health Insurance Regulation:  Variation in
Recent State Small Employer Health Insurance Reforms
(GAO/HEHS-95-161FS, June 12, 1995).  While some medium firms'
premiums are experience-rated, other firms in this size group are
charged premiums that are based on a combination of the claims cost
history of their own employees and the claims cost experience of a
much larger pool of individuals. 


   GAPS IN ELIGIBILITY LIMIT MANY
   EMPLOYEES' ACCESS TO HEALTH
   INSURANCE FROM THEIR EMPLOYERS
------------------------------------------------------------ Letter :5

Although most medium and large businesses offer health coverage, many
employees lack access to the health benefits their employers sponsor. 
In particular, part-time employees or those with a temporary work
arrangement are not eligible for their employer-sponsored health
insurance plans.\19 Although more than half of such employees obtain
health coverage from other sources, such as through their spouses,
Medicare or Medicaid, or the individual insurance market, the
uninsurance rates for part-time and temporary employees are high--25
and 33 percent, respectively. 


--------------------
\19 We use the Bureau of Labor Statistics' (BLS) definition of
part-time employees, which applies to persons who work less than 35
hours per week.  However, employers vary in how they define part-time
work.  For example, an employee who works 32 hours may be considered
to be a full-time worker in firm A; this same employee may be
considered a part-time employee in firm B.  We use the term temporary
work to refer to any job that is structured to be of limited
duration.  Temporary employees are individuals in jobs that are
expected to last only a limited period of time.  This definition
includes wage and salary, self-employed, and independent workers who
expect their jobs to end within a year or do not expect the jobs to
last.  BLS classifies these employees as contingent workers. 


      MOST PART-TIME WORKERS
      INELIGIBLE FOR
      EMPLOYMENT-BASED HEALTH
      COVERAGE RELY ON OTHER
      SOURCES BUT MANY ARE
      UNINSURED
---------------------------------------------------------- Letter :5.1

Part-time employees, who constitute about 19 percent of the total
workforce, are much less likely to be eligible for health insurance
than full-time employees, even though their employers offer coverage. 
As table 3 shows, in 1997, 31 percent of part-time employees were
eligible for their employer-sponsored health insurance compared with
82 percent of full-time employees.  Moreover, the percentage of firms
in which part-time employees were eligible for coverage declined from
55 percent in 1994 to 47 percent in 1997, according to KPMG Peat
Marwick data. 



                          Table 3
          
             Part-Time and Full-Time Employees'
           Health Insurance Status, February 1997

                                 Part-     Full-
Health insurance status           time      time       All
----------------------------  --------  --------  --------
Eligible for coverage from         31%       82%       72%
 current employer health
 plan
Total covered                      75%       84%       82%
Through current employer            17        70        59
Through other sources\a             58        14        23
Uninsured                          25%       16%       18%
Total number employed aged      21,297    90,268  111,837\
 16 years and older (in                                  b
 thousands)
----------------------------------------------------------
\a Other sources include other job, union, spouse's health plan,
Medicaid, Medicare, TRICARE, or purchased in the individual insurance
market. 

\b The total is greater than the sum of part-time and full-time
employees because the usual status of a person's main job could not
be determined for a small number of respondents. 

Source:  BLS, unpublished tabulations from the February 1997
Contingent and Alternative Employment Arrangement Survey. 

Despite their relatively low eligibility rate, about three-quarters
of part-time employees had health insurance coverage in 1997.  The
main reason for the discrepancy between eligibility and coverage
rates is that over half of part-time employees were able to obtain
health insurance coverage through other sources, such as other jobs,
family members, the individual insurance market, and Medicare or
Medicaid.  Nonetheless, about one-fourth of the part-time workforce
(or 5.4 million employees) was uninsured as of February 1997,
according to data from BLS.  Uninsured part-time employees accounted
for about 27 percent of the 19.9 million uninsured wage and salary
workers in 1997.  The majority of the uninsured part-time employees
were between the ages of 16 and 34, a group that generally has lower
earnings compared with most older workers.\20

Similar disparities between the eligibility of part-time and
full-time workers are observed when looking at data for different
firm sizes.  Agency for Health Care Policy and Research (AHCPR) data
for 1996 show that full-time employees were offered health insurance
from their current main employers at more than twice the rate of
part-time employees for firms with up to 200 employees.  For firms
with more than 200 workers, the ratio of full-time to part-time rates
of eligibility is about 1.7 to 1.  (See table 4.)



                          Table 4
          
             Part-Time and Full-Time Employees
          Offered Health Insurance at Current Main
                 Job by Employer Size, 1996

                    (Numbers in percent)

                                 Part-     Full-
Number of employees               time      time       All
----------------------------  --------  --------  --------
Up to 50                            19        66        53
51-100                              40        90        80
101-200                             37        91        83
Over 200                            57        96        92
All                                 27        80        69
----------------------------------------------------------
Note:  Employer size is defined by the number of employees working
for organizations.  Percentages include private and public employees. 
Public sector employment accounted for about 16 percent of U.S. 
workers aged 16 years and over in 1996. 

Source:  HHS/AHCPR Medical Expenditure Panel Survey (MEPS). 


--------------------
\20 Like uninsured part-time workers, the uninsured in general tend
to be young. 


      TEMPORARY EMPLOYEES ARE LESS
      LIKELY TO BE OFFERED HEALTH
      INSURANCE THAN PERMANENT
      EMPLOYEES
---------------------------------------------------------- Letter :5.2

In addition to work schedule, anticipated job duration--short-term or
temporary versus more long-term or permanent--affects the likelihood
that employees will be eligible for their employer-provided health
benefits.  Temporary employees are less than half as likely to be
eligible for health coverage than permanent workers.  Specifically,
of the 5.1 million temporary employees in the U.S.  workforce in
1997, only 35 percent were eligible to participate in their
employers' health insurance.\21 In contrast, 74 percent of permanent
employees were eligible for their employers' health insurance. 

The percentage of temporary workers eligible for their employers'
health insurance varies across industries.  Private-sector industries
with the lowest eligibility rates for temporary employees in 1997
were agriculture (18 percent), retail (23 percent), and construction
(29 percent).  By contrast, eligibility rates for temporary employees
were highest among the manufacturing and transportation industries
(45 and 46 percent).  In general, the data show a direct relationship
between the eligibility rates for temporary workers and the offer
rates for all employees within a particular industry sector.  For
example, industries characterized by low rates of eligibility for
temporary workers (agriculture, retail trade, and construction) also
had relatively low offer rates for all employees, as shown in figure
3. 

Although eligibility rates for temporary workers were lower than for
permanent workers, two-thirds of temporary workers had health
insurance from a variety of sources (often through family health
coverage).  Nonetheless, the uninsurance rate for temporary workers
was about 33 percent (or 1.7 million people) in 1997 compared with 17
percent for permanent workers.  Moreover, temporary workers generally
earn less than permanent workers.  The median weekly earnings of
temporary workers in 1997 was $266 compared with $444 for permanent
workers.\22 Consequently, many temporary workers might not be able to
afford their share of health insurance cost even if they were
eligible for their employer-sponsored health plans. 


--------------------
\21 Temporary employees constituted about 5 percent of the U.S. 
workforce in 1997.  Data are not readily available to categorize
temporary workers by firm size; thus, this discussion compares
eligibility rates for all temporary workers with all permanent
workers. 

\22 The lower median earnings for temporary workers is influenced by
the relatively high concentration of part-time workers in that
category.  In 1997, 43 percent of temporary workers were part-time,
compared with 18 percent of permanent workers. 


   CONCLUSIONS
------------------------------------------------------------ Letter :6

Under HIPAA, insurers cannot deny coverage to a firm with 50 workers
or less, but some have questioned whether firms just past this
threshold escape the difficulties that small firms used to face.  As
the evidence presented above shows, virtually all medium and large
firms have access to health insurance.  However, health coverage is
not universal.  Despite access to insurance, about 10 percent of
firms with 50 to 99 workers (and a small proportion of larger firms)
choose not to offer it, and some firms may be deterred by what they
see as unfavorable or unaffordable premiums.  Perhaps more important,
even when an employer offers health insurance, the offer may not
extend to all the firm's workers.  Part-time and temporary employees
often are not eligible for an employer's plan, and lower-wage workers
may not find it affordable.  In the market for health insurance,
employers' and workers' decisions are voluntary.  Consequently, a
reversal of an employer's decision to forgo offering coverage or an
employee's decision to decline an offer hinges on coverage becoming
more attractive and affordable. 


   COMMENTS FROM OUTSIDE REVIEWERS
------------------------------------------------------------ Letter :7

We obtained comments on a draft of this report from a number of
experts on private health insurance including those with NCHS, the
Employee Benefit Research Institute, and the Health Insurance
Association of America.  The reviewers agreed with the contents of
the report and suggested a number of technical corrections that we
incorporated where appropriate. 


---------------------------------------------------------- Letter :7.1

We are sending copies of this report to interested congressional
committees and are making copies available to others on request. 

Please call me at (202) 512-7114 if you or your staff have any
questions about this report.  Other GAO contacts and staff
acknowledgments for this report are listed in appendix III. 

William J.  Scanlon
Director, Health Financing and
 Systems Issues


List of Addressees

The Honorable William V.  Roth, Jr.
Chairman
The Honorable Daniel Patrick Moynihan
Ranking Minority Member
Committee on Finance
United States Senate

The Honorable James M.  Jeffords
Chairman
The Honorable Edward M.  Kennedy
Ranking Minority Member
Committee on Labor and Human Resources
United States Senate

The Honorable Thomas J.  Bliley, Jr.
Chairman
The Honorable John D.  Dingell
Ranking Minority Member
Committee on Commerce
House of Representatives

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

The Honorable William Archer
Chairman
The Honorable Charles B.  Rangel
Ranking Minority Member
Committee on Ways and Means
House of Representatives


METHODOLOGY
=========================================================== Appendix I

The Health Insurance Portability and Accountability Act of 1996 (P.L. 
104-191, Aug.  21, 1996) directed us to examine the extent to which
classes of large employers in the different states have access to
health insurance and the circumstances for the lack of access, if
any.  In consultation with the offices of the committees interested
in the subject matter, we agreed to report on (1) the extent to which
medium and large employers in different states and categories have
access to health insurance and the barriers (if any) that these
employers face in obtaining health insurance; (2) major factors,
including health insurance costs, that affect employers' decisions to
offer health insurance; and (3) the extent to which employees are
eligible for and covered by their employers' health insurance.\23


--------------------
\23 While the act speaks only of small and large employers and
defines large employers as those that employed an average of at least
51 employees during the preceding calendar year and that employ at
least 2 employees on the first day of a plan year, we are using the
terms medium and large employers to emphasize the broad range of
employers included in that definition--firms employing from 51 to
several thousand employees. 


   MEDIUM AND LARGE EMPLOYERS'
   ACCESS TO HEALTH INSURANCE
--------------------------------------------------------- Appendix I:1

Because direct measures of access do not exist, we used available
data on employers that offer health insurance as a proxy for the
extent of access.  This may understate the true level of employers'
access to insurance because it excludes employers that have access to
health insurance but decide not to offer it to their employees. 

To analyze employers' access to health insurance in different states
and classes, we used data covering 1993 (the most recent source for
state-level estimates) from the National Employer Health Insurance
Survey (NEHIS) and from KPMG Peat Marwick's survey of employers for
1996. 


      NEHIS
------------------------------------------------------- Appendix I:1.1

NEHIS was managed by the National Center for Health Statistics
(NCHS), a component of the Centers for Disease Control and Prevention
of the Department of Health and Human Services in partnership with
the Agency for Health Care Policy Research and the Health Care
Financing Administration.  This survey had usable responses from
about 35,000 private business establishments.  The sampling unit for
NEHIS was the establishment, defined as "an economic unit, generally
at a single physical location, where business is conducted or where
services or industrial operations are performed."\24 A primary reason
that NEHIS sampled establishments rather than firms is that
establishments are confined within state borders, allowing state
estimates. 

NEHIS included a probability sample of employers in each state and
the District of Columbia, and we used these data because the data set
was large enough to allow us to

  -- group establishments into categories by number of employees,
     such as firms with 50 or fewer employees, 51 to 100 employees,
     101 to 200 employees, and more than 200 employees,\25

  -- report state-level information; and

  -- group establishments into categories, such as type of industry. 

Businesses in all states and the District of Columbia are represented
in the survey.  The minimum number of participating establishments
was 383 in Alaska; the maximum number was 1,083 in California. 
Respondents were asked to describe the characteristics of their firms
and their health insurance programs as of December 31, 1993.  The
results from NEHIS included in this report are for private sector
establishments only. 

NCHS staff performed certain data analyses for us at our request.  We
did not independently verify the data entry or data analysis work
NCHS performed, although we confirmed that the results of its
analyses for us were consistent with results contained in a report
NCHS prepared under its own name. 

Survey data based on samples of populations are subject to sampling
errors, which can be expressed as confidence intervals around a point
estimate.  The estimated percentage of establishments that offered
health insurance in the states, the District of Columbia, and
nationwide are shown with the corresponding confidence intervals for
those point estimates in appendix II.  Those data are depicted as
bars, showing the relative position of the estimated offer rates and
the confidence interval for those estimates at the 95-percent
confidence level.  NEHIS data were also used in other tables
describing features of establishments that offered health insurance. 
The estimated sampling errors (at the 95-percent confidence level)
for the percentage of establishments offering health insurance are

  -- by industry type and firm size, in figure 3, plus or minus 1
     percent to plus or minus 9 percent and

  -- by general wage level, by firm size and nationwide, in table 2,
     plus or minus 1 percent to plus or minus 8 percent. 


--------------------
\24 Establishments are not the same as firms.  Firms may be composed
of one or more establishments under common ownership or control, and
a large firm may have several establishments (or locations) where
economic activity takes place within a single state or in several
states. 

\25 For grouping by firm size, the unit of analysis is the
establishment.  Thus, survey responses from an establishment of 50
employees that is part of a firm with more than 200 employees
nationwide would be included in the group of firms with more than 200
employees. 


      KPMG PEAT MARWICK
------------------------------------------------------- Appendix I:1.2

KPMG Peat Marwick conducts and publishes its results from surveys on
employee compensation and benefits.  We used data from KPMG Peat
Marwick (some published and some unpublished) to provide more current
estimates of the percentage of business firms in various size
categories that offered health insurance in 1996.  These survey data
included the company's annual survey of firms employing 200 or more
employees, supplemented by an additional survey of firms with fewer
than 200 employees.  The surveys were stratified random surveys of a
total of about 2,600 employers nationwide.  While more current than
the 1993 NEHIS data, KPMG Peat Marwick did not report state-level
estimates from its survey.  Thus, we used these data to compare with
the NEHIS data at the national level.\26 We did not independently
verify the KPMG Peat Marwick data, but these data are widely used by
researchers. 


--------------------
\26 The data needed to compute sampling errors for the estimates from
KPMG Peat Marwick were not available to us. 


   FACTORS INVOLVED IN EMPLOYER
   DECISIONS TO OFFER HEALTH
   INSURANCE
--------------------------------------------------------- Appendix I:2

To examine the reasons that employers who had access to insurance
decided not to offer health insurance to their employees, we reviewed
recent studies on health insurance published by health insurance
researchers and our earlier reports and interviewed a broad range of
people who were knowledgeable about the insurance market and
employers' behavior.  We interviewed benefit consultants and health
policy researchers; representatives of employers in California,
Pennsylvania, and Virginia who employed from 80 to 185 employees;
insurance agents and brokers in California, Maine, Massachusetts, and
Oregon; representatives of the Health Insurance Association of
America and insurance companies; and state insurance regulators in
Oregon, Pennsylvania, Utah, and Wisconsin.  We selected the persons
we interviewed to provide a geographical cross-section of the nation. 


   EMPLOYEE ACCESS TO
   EMPLOYER-SPONSORED HEALTH
   INSURANCE
--------------------------------------------------------- Appendix I:3

Another aspect of access to health insurance is the employee's view. 
Both an employer and employee may have access to insurance.  The
employer may decide to offer insurance, but an employee may decline
to enroll.  People may decline to enroll for a variety of reasons,
including not being able to afford the insurance or having coverage
through another source, such as another family member's coverage. 

Our data sources for employee access to health insurance included

  -- the Agency for Health Care Policy and Research and NCHS' Medical
     Expenditure Panel Survey (MEPS) Household Component, which
     provides current information on health insurance offers and
     coverage from a sample of about 25,000 persons, and

  -- the Bureau of Labor Statistics' (BLS) 1994, 1995, and 1997
     surveys of small, medium, and large employers. 

The MEPS data are the source of our estimate of the percentage of
employees eligible for their employer-based health insurance plans in
1996, by firm size and work schedule, included in table 4.  The
sampling errors for those estimates are plus or minus 1 percent to
plus or minus 8 percent.  Estimates of sampling errors for data we
obtained from BLS were not available to us.  Although we did not
independently verify the accuracy of the data, the MEPS and BLS data
that we used were published or made available for others' use by the
sponsoring agencies. 


HEALTH INSURANCE OFFER RATES BY
FIRM SIZE AND STATE, 1993
========================================================== Appendix II

   Figure II.1:  Health Insurance
   Offer Rates for Firms With Up
   to 50 Employees and Firms With
   More Than 50 Employees,
   Nationwide and by State, 1993,
   Based on NEHIS Data

   (See figure in printed
   edition.)

   Figure II.2:  Health Insurance
   Offer Rates for Firms With 51
   to 100 Employees, Nationwide
   and by State, 1993, Based on
   NEHIS Data

   (See figure in printed
   edition.)

   Figure II.3:  Health Insurance
   Offer Rates for Firms With 101
   to 200 Employees, Nationwide
   and by State, 1993, Based on
   NEHIS Data

   (See figure in printed
   edition.)

   Figure II.4:  Health Insurance
   Offer Rates for Firms With More
   Than 200 Employees, Nationwide
   and by State, 1993, Based on
   NEHIS Data

   (See figure in printed
   edition.)


GAO CONTACTS AND STAFF
ACKNOWLEDGMENTS
========================================================= Appendix III

GAO CONTACTS

Jonathan Ratner, Senior Health Economist, (202) 512-7107
Carmen Rivera-Lowitt, Evaluator-in-Charge, (202) 512-4342

STAFF ACKNOWLEDGMENTS

Roger Hultgren, Senior GAO Evaluator, assisted in the design and
implementation of the study.  Paula Bonin, C.  Robert DeRoy, and
Elsie Picyk provided analyses of computerized databases. 

RELATED GAO PRODUCTS

Private Health Insurance:  Declining Employer Coverage May Affect
Access for 55- to 64-Year-Olds (GAO/HEHS-98-133, June 1, 1998). 

Health Insurance Standards:  Implications of New Federal Law for
Consumers, Insurers, Regulators (GAO/T-HEHS-98-114, Mar.  19, 1998). 

Health Insurance Standards:  New Federal Law Creates Challenges for
Consumers, Insurers, Regulators (GAO/HEHS-98-67, Feb.  25, 1998). 

The Health Insurance Portability and Accountability Act of 1996: 
Early Implementation Concerns (GAO/HEHS-97-200R, Sept.  2, 1997). 

Private Health Insurance:  Continued Erosion of Coverage Linked to
Cost Pressures (GAO/HEHS-97-122, July 24, 1997). 

Retiree Health Insurance:  Erosion in Employer-Based Health Benefits
for Early Retirees (GAO/HEHS-97-150, July 11, 1997). 

Employment-Based Health Insurance:  Costs Increase and Family
Coverage Decreases (GAO/HEHS-97-35, Feb.  24, 1997). 

Private Health Insurance:  Millions Relying on Individual Market Face
Cost and Coverage Tradeoffs (GAO/HEHS-97-8, Nov.  25, 1996). 

Health Insurance Regulation:  Varying State Requirements Affect Cost
of Insurance (GAO/HEHS-96-161, Aug.  19, 1996). 

Health Insurance Portability:  Reform Could Ensure Continued Coverage
for Up to 25 Million Americans (GAO/HEHS-95-257, Sept.  19, 1995). 

Employer-Based Health Plans:  Issues, Trends, and Challenges Posed by
ERISA (GAO/HEHS-95-167, July 25, 1995). 

Health Insurance Regulation:  Variation in Recent State Small
Employer Health Insurance Reforms (GAO/HEHS-95-161FS, June 12, 1995). 

Employer-Based Health Insurance:  High Costs, Wide Variation Threaten
System (GAO/HRD-92-125, Sept.  22, 1992). 


*** End of document. ***