XI. BUILDING HEALTH AND SAFETY INTO
EMPLOYMENT RELATIONSHIPS IN THE
CONSTRUCTION INDUSTRY

The Construction Industry and
Uncompensated Health Care

C. Jeffrey Waddoups
University of Nevada, Las Vegas

Abstract

     The objective of this study is to illuminate the relationship between employment-based health insurance and uncompensated health care costs, with particular emphasis on the construction industry and on differences between non-Hispanic and Hispanic workers. The findings suggest that the organization of work in the construction industry leads to comparatively low rates of employment-based health insurance, especially among non-union workers and those reporting Hispanic ethnicity. Furthermore, data from a major urban hospital and its constellation of clinics shows that the lack of insurance coincides with a disproportionate use of uncompensated health care among construction workers. Such a disproportionate use of uncompensated care is a concrete example of how employers in industries such as construction pass costs of health care onto communities and their workers.

Introduction

     Workers in the construction industry are at greater risk of occupational injuries and deaths than their counterparts in other industries. The incidence of on-the-job fatalities is particularly severe among Hispanic workers (Center to Protect Workers' Rights [CPWR] 2002, pp. 33-34). Not only is construction work more dangerous than work in other industries, but the incidence of employer-based health insurance (EBHI) is generally lower as well, especially for Hispanic workers. The preponderance of the evidence suggests that individuals with inadequate health insurance coverage experience poorer health outcomes than their insured counterparts (e.g. Hadley 2002).

     Besides personal costs borne by construction workers and their families from dangerous work and inadequate access to health care financing, the community also incurs costs when uninsured workers and their dependents consume health care that remains uncompensated and must be subsidized from other sources. Such costs are manifested through higher taxes, higher prices of health care to paying patients, higher-priced insurance premiums, and, because employers treat EBHI as a cost of doing business, through a combination of higher prices, lower wages, and diminished profits. Thus, the community can be thought of as subsidizing low-wage employers that do not offer practically affordable EBHI.

     The objective of this study is to illuminate the relationship between EBHI and uncompensated health care costs to a community, with particular emphasis on the construction industry and on differences between non-Hispanic and Hispanic workers. To study the issues, I focus on data drawn from Clark County, Nevada. The area consists of Las Vegas and surrounding municipalities, with a population of nearly 1.5 million. Population in Nevada increased by 66.3 percent between 1990 and 2000 (U.S. Bureau of the Census 2002). Not surprisingly, the growth has been accompanied by a relatively large concentration of employment in construction. Mirroring an emerging national trend, Hispanic workers are proportionately more likely to be employed in the industry than their non-Hispanic counterparts (CPWR 2002, p. 16).

     The health care infrastructure includes a public hospital with a complement of clinics, which are the region's safety-net health care providers and deliver a majority of uncompensated health care in the area. Although the study focuses on a single community, the literature on EBHI and health care finance suggests that the issues addressed in the article are common to many communities in the United States.

The Incidence of EBHI by Industry and Ethnicity

     Universally held EBHI would render uncompensated health care costs

among the employed a nonissue, but EBHI is not universal among the employed or their dependents. I used data from the Current Population Survey (CPS), March supplement, to estimate the incidence of health insurance coverage. The CPS is a common data source for making such estimates. In the March CPS, respondents provide information on their employment and health insurance status. If respondents report coverage through their own or a relative's employer or union, it is recorded as "employment based." Other possibilities include privately purchased insurance or insurance obtained through government-sponsored programs, such as Medicaid, Medicare, the military, and/or the Veterans Administration.

     Although the CPS data can be used to estimate the proportion of residents with health insurance at the state level, the sample design does not ensure unbiased estimates for smaller geographic areas such as counties. Thus, I report the incidence of health insurance coverage at the state and county levels, keeping in mind that the county results must be treated with some caution.

     Another consideration for the precision of the estimates is sample size. The State Health Access Data Assistance Center (SHADAC 2001) suggests combining three years of state level CPS data to obtain sufficiently large samples. Following SHADAC, I estimated the incidence of health insurance by location, industry, and Hispanic ethnicity, using 1998-2000 CPS data. The results are presented in Table 1.

     Table 1 reveals substantial differences in EBHI and the category of "any coverage" by industry and ethnicity. Non-Hispanic respondents in the construction industry are covered by EBHI at a similar rate to "trade" at the state level and "services" at the county level. Rates of EBHI for Hispanics are uniformly lower in construction and other major industries except the hotel industry, which is highly unionized in southern Nevada (Waddoups 1999). If the consequence of low health insurance coverage rates is poorer health outcomes, then construction workers (and their dependents), and in particular those of Hispanic ethnicity, appear to face a relatively higher risk of such negative outcomes.

Distribution of Uncompensated Health Care Costs by
Industry and Ethnicity

     One would expect workers employed in industries with lower rates of EBHI to have fewer financial resources to finance health care and thus to consume a disproportionately large share of uncompensated care. I tested the expectation using the safety-net provider's account data on patients and their guarantors (i.e., those responsible for payment). When treatment is received, officials in the hospital or clinics record a guarantor's employment status, which I used to allocate accounts into industry categories. I then computed expected uncompensated care costs for each industry, assuming uncompensated care account balances are proportional to employment shares of the industries. Next, I estimated the actual uncompensated care costs attributable to each industry. Finally, I calculated the deviation of expected from actual costs to reveal the industries that contribute disproportionately large (or small) shares to uncompensated care costs.

     Briefly referring back to Table 1, the low incidence of EBHI among workers in construction would lead one to expect the industry to be overrepresented in uncompensated care consumption. Indeed, results in Table 2 indicate that both non-Hispanic and Hispanic guarantors contribute a disproportionate share to uncompensated care costs. Non-Hispanic guarantors employed in construction contribute 81 percent more to uncompensated care costs than expected, given their share of employment. The figure for Hispanic guarantors is 90.4 percent. The larger figure for Hispanic workers corresponds to a lower incidence of insurance coverage as reported in Table 1.

     Combining the findings in Tables 1 and 2, one can clearly see the relationship between the lack of EBHI and the disproportionate representation in the uncompensated care categories. The disproportionate representation is a concrete example of how employers in the construction sector are particularly likely to employ workers who use uncompensated health care at the public's expense. It also demonstrates how the community directly subsidizes employers who do not provide practical access to EBHI for its workers.

Organization of Work, EBHI, and Collective Bargaining

     The organization of work in construction, combined with the U.S. system of EBHI for providing health care financing, makes workers in the construction industry less likely to be covered than workers in most other industries. Previous research has shown that workers in small firms and who change employers frequently are among those least likely to have health insurance through work (Henderson 1999). The construction industry is disproportionately composed of small contractors and is characterized by an organization of work that practically ensures that most workers will change employers frequently (Grob 1994). Thus, institutional characteristics of the industry tend to restrict construction workers' access to EBHI.

     Collective bargaining, however, reduces the importance of such obstacles by tying eligibility for EBHI to a union rather than an employer. Thus, among unionized construction workers, employment in small firms and frequent changes of employer are less likely to prevent access to EBHI (e.g., Petersen 2000). A challenge to reliance on collective bargaining to increase EBHI in construction is low union density. Although the union density is higher than in most industries, only 20 percent of workers are covered by collective bargaining contracts (Hirsch and MacPherson 2002).

     To evaluate the potential impact of collective bargaining on EBHI coverage, I used national CPS data from the years 1998-2000 to estimate the probabilities that nonmanagerial, nonprofessional/technical workers in five major industry categories were covered by EBHI. The results in Table 3 are computed from logistic regression models. They show that workers covered by a union contract are more likely to have EBHI, which has been demonstrated elsewhere (e.g., Wiatrowski 1994). Perhaps more surprising, however, is the difference in rates of EBHI in construction by union status compared to differences in other industries. Unionized construction workers are covered by EBHI at a rate of .894, whereas the rate for nonunion workers is only .615. Trade has the next largest gap, with .842 and .729 for union and nonunion workers.

     It appears that institutional obstacles to obtaining EBHI affect construction workers to a greater degree than workers in other industries and that collective bargaining could be an effective mechanism to break down such obstacles. The results in Table 3 demonstrate that health insurance coverage among unionized construction workers approaches the higher rates found in other industries. By inference, one may conclude that the high rates of uncompensated care attributable to the construction sector probably originate from the nonunion sector of the industry because of its lower incidence of EBHI. One may also conclude that rates of uncompensated care could be substantially reduced if collective bargaining became more prevalent.

Conclusion

     The findings suggest that the organization of work in the construction industry leads to comparatively low rates of EBHI, especially among nonunion workers and those reporting Hispanic ethnicity. Low rates of coverage, in turn, appear to result in disproportionately high levels of uncompensated health care costs, which are eventually financed by the community.

     In place of traditional EBHI, the community is providing access to health care through its public hospital and clinics. It is not, however, the kind of access that necessarily leads to the more positive health outcomes that have been found to occur among those covered by health insurance. Uninsured workers and their dependents treated at the public hospital and clinics are billed for services rendered. Thus, even partial payments to cover bills often lead to financial hardship. Rather than face the prospect of financially ruinous bills for health care, treatment is often delayed until the health event has either resolved itself or reached emergency status. Emergency treatment provided to the uninsured is often quite expensive and is thus likely to become uncompensated care. Among the employed, it appears that workers in construction (and their dependents) are disproportionately subject to such financial and health insecurity.

     Although I have focused the present study on the experience of the construction industry in southern Nevada, workers, safety-net health care providers, and communities across the United States face similar issues. It appears that the present system of voluntary EBHI as the cornerstone of health care financing does not provide a subsistence level of health care for many workers in construction and other industries. Furthermore, one may envision a number of plausible scenarios that could further destabilize the already precarious positions of workers and safety-net health care providers. For example, consider the consequences of a significant downturn in economic activity that reduces the incidence of EBHI, an increased reluctance of decision makers and taxpayers to subsidize costs of uncompensated health care, or a continued decline in union density.

Acknowledgment

     This research is supported under grant U60/CCU317202 through a cooperative agreement with the Center to Protect Workers' Rights and the National Institute of Occupational Safety and Health.

References

Grob, Heather. 1994. "Developments in Collective Bargaining in Construction in the 1980s and 1990s." In Paula Voos, ed., Contemporary Collective Bargaining in the Private Sector. Madison, WI: Industrial Relations Research Association, pp. 411-45.

Hadley, Jack. 2002. Sicker and Poorer: The Consequences of Being Uninsured. Report prepared for the Kaiser Commission on Medicaid and the Uninsured. Washington, D.C.: The Urban Institute.

Henderson, James W. 1999. Health Economics and Policy. Cincinnatti, OH: Southwestern College Publishing.

Hirsch, Barry, and David A. MacPherson. 2002. "Unionstats.com: Union Coverage and Membership Database" [www.trinity.edu/bhirsch; October 16, 2002].

Petersen, Jeffrey S. 2000. "Health Care and Pension Benefits for Construction Workers: The Role of Prevailing Wage Laws." Industrial Relations, Vol. 39, no.2, pp. 246-64.

State Health Access Data Assistance Center. 2001. "The Current Population Survey and State Health Insurance Estimates." Issue Briefs, Issue 1 (March), pp. 1-4 [http://www.shadac.umn.edu/publications/docs/IB-01.pdf].

The Center to Protect Workers' Rights. 1998. The Construction Chart Book: The U.S. Construction Industry and its Workers. Washington D.C.: The Center to Protect Workers Rights.

United States Census Bureau. 2002. State and County Quick Facts [http://quickfacts.census.gov/qfd/states/32000.html; November 14, 2002].

Waddoups, C. Jeffrey. 1999. "Union Wage Effects in Nevada's Hotel and Casino Industries." Industrial Relations, Vol. 38, no. 4, pp. 577-83.

Wiatrowski, William J. 1994. " Employee Benefits for Union and Nonunion Workers." Monthly Labor Review, Vol. 117, no. 2 (February), pp. 34-7.