What GAO Found
The five studies GAO reviewed that used microsimulation models to estimate the effects of the Patient Protection and Affordable Care Act (PPACA) on employer-sponsored coverage generally predicted little change in prevalence in the near term, while results of employer surveys varied more widely. The five microsimulation study estimates ranged from a net decrease of 2.5 percent to a net increase of 2.7 percent in the total number of individuals with employer-sponsored coverage within the first 2 years of implementation of key PPACA provisions, affecting up to about 4 million individuals. Two of these studies also indicated that the majority of individuals losing employer-sponsored coverage would transition to other sources of coverage. In contrast to the microsimulation studies, which estimate the net effect on individuals, most employer surveys measure the percentage of employers that may drop coverage in response to PPACA. Among the 19 surveys, 16 reported estimates of employers dropping coverage for all employee types. Among these 16, 11 indicated that 10 percent or fewer employers were likely to drop coverage in the near term, but estimates ranged from 2 to 20 percent. Most surveys were of employers currently offering coverage and therefore did not also address whether other employers may begin to offer coverage in response to PPACA; however, 3 that did found that between 1 and 28 percent would begin offering coverage as a result of PPACA. Longer-term predictions of prevalence of employer-sponsored coverage were fewer and more uncertain, and four microsimulation studies estimated that from about 2 million to 6 million fewer individuals would have employer-sponsored coverage in the absence of the individual mandate compared to with the mandate.
Differences in key assumptions and consideration of PPACA provisions likely contributed to some variation among estimates from the five microsimulation studies and the 16 employer surveys. Variation among the microsimulation studies may have stemmed from differences in assumptions about employer and employee decision making, the time frames of the estimates, and assessments of potential compliance with the individual mandate. Variation among the employer surveys may be related to differences in survey sampling techniques, the number and types of employer respondents, and the framing of survey questions. For example, some surveys used a random sampling methodology, allowing their results to be generalized across all employers, while others did not. Also, some referred to specific PPACA provisions or provided specific information about provisions to respondents, while others did not.
Some of the 19 employer surveys indicated that PPACA may have a larger effect on small employers and certain populations and may prompt some employers to change benefit designs. For example, 4 surveys found that smaller employers were more likely than other employers to stop offering health coverage in response to PPACA, and 5 found that employers in general were more likely to drop coverage for retirees than for all employees. Nine surveys also indicated that employers are considering key changes to benefit design, some of which may result in greater employee cost for health coverage.
GAO provided a draft of this report to two researchers with expertise in employee health benefits issues. The experts agreed with GAOs report and provided technical comments, which were incorporated as appropriate.
Why GAO Did This Study
The share of employers offering health coverage has generally declined in the last decade. Researchers believe that certain provisions of PPACA could affect employers future willingness to offer health coverage, such as the availability of subsidized coverage through new health insurance marketplaces called exchanges and an individual mandate, which will require most people to obtain health coverage or pay a tax penalty. Certain PPACA provisions are scheduled to take effect in 2014. Researchers have provided various estimates of the effect PPACA may have on employer-sponsored coverage.
GAO was asked to review the research on this topic. GAO examined (1) estimates of the effect of PPACA on the extent of employer-sponsored coverage; (2) factors that may contribute to the variation in estimates; and (3) how estimates of coverage vary by the types of employers and employees that may be affected, as well as other changes employers may be considering to the health benefits they offer. GAO reviewed studies published from January 1, 2009, through March 30, 2012 containing an original numerical estimate of the prevalence of employer-sponsored coverage at the national level. These included 5 microsimulation models and 19 employer surveys. Microsimulation models can systematically estimate the combined effects of multiple PPACA provisions in terms of both gains and losses of coverage; their results are based on multiple data sets and assumptions. Surveys reflect employer perspectives; they have limits as a predictive tool in part based on varied survey methodologies and respondent knowledge of PPACA.
For more information, contact John E.Dicken at (202) 512-7114 or firstname.lastname@example.org.