Employer-sponsored insurance (ESI) plays a critical role in the health care market in the United States. In 2017, 56% of the population was covered through an ESI plan. The link between employment and health insurance coverage is largely the result of the fact that ESI is a non-taxed fringe benefit. This link may affect individuals’ labor market behavior. There are costs associated with non-portability of health coverage when switching jobs or leaving employment. These costs could be high if health coverage is denied due to preexisting conditions, or relatively small, if an individual changes health care networks. An individual may stay in a job to retain health coverage to avoid these costs, which is a phenomenon known as “job-lock.” Previous studies show that ESI reduces job mobility since health insurance is offered as some portion of compensation for these workers. This effect was found to be salient among individuals with chronic health conditions. Job-lock could be an important concern if workers are unable to match with productivity-enhancing jobs.
The Patient Protection and Affordable Care Act (ACA), or commonly called “Obamacare,” was enacted in March 2010. The ACA altered the health care system through four key provisions: i) the expansion of Medicaid; ii) the individual mandate; iii) the creation of health insurance exchanges; and iv) the employer mandate. Each of these provisions can affect employment and the composition of employee compensation. The Medicaid expansion was made optional for the states through the Supreme Court’s 2012 ruling and the individual mandate will no longer be in effect next year.
I have studied how the differential expansion of Medicaid across the states affected the labor supply of older workers. The graph below depicts the shares of the US population that are covered by Medicaid and are uninsured for the years 2008 to 2017.
Source: Author’s calculations, 2008 to 2017 American Community Survey, 1-Year Estimates
The enactment of Medicaid, a means-tested health insurance program, dates to 1965. Its purpose is to provide coverage to low-income individuals that fit into categorical eligibility, some of which are the disabled, mentally ill, children, and pregnant women. The ACA eliminated eligibility categories for the poor, and expanded Medicaid to all Americans up to age 65 with incomes under 138% of the federal poverty level (FPL). The 2014 Medicaid expansion was perhaps the biggest change under the ACA. Some studies have found a moderate Medicaid crowd-out of private health insurance among low-income individuals and those who are aged 55-64. Note that, in June 2013, more than 40% of individuals below 250% of the FPL had ESI.
The ACA’s Medicaid expansion reduced costs regarding non-portability of health coverage for individuals with ESI. Since there are less costs associated with changing jobs or leaving employment with respect to the expansion, one could expect a decline in the job-lock phenomenon. It is also possible that individuals may manipulate their working hours to keep their income below the eligibility threshold. In a recent study, I look at the early retirement decision of childless adults aged 55-64 by exploiting the expansion decision of states resulting from the 2012 Supreme Court ruling and the timing of expansions under the ACA (see the Map below). I find that Medicaid enrollment increases for both men and women after the expansion. I also show that the Medicaid expansions do not affect the retirement decision of men, but women were more likely to retire early. This interesting finding points out that labor supply decisions may not only depend on one’s own health insurance, but also on a spouse’s health insurance and labor market responses.
Source: Author’s calculations, Status of State Action on the Medicaid Expansion Decision, Kaiser Family Foundation
The ACA also created a market for private insurers, known as the health insurance exchanges (or the Marketplace), to offer different individual and family plans. The “metal” plans differ based on actuarial value and come in four different tiers: bronze plans cover 60% of out-of-pocket (OOP) expenses; silver plans cover 70% of OOP expenses; gold plans cover 80% of OOP expenses; and platinum plans cover 90% of OOP expenses. To incentivize purchases from the exchanges, the ACA provided premium tax credits and cost-sharing subsidies for individuals purchasing in the marketplaces with incomes between 100% and 400% FPL. Additionally, the regulations prohibited insurers from denying coverage for individuals with preexisting conditions, referred to as “guaranteed issue,” and restricted premiums to vary only based on age, smoking status, geographic rating area, and family composition. The subsidized family plan options, paired with guaranteed issue, may serve as a source of portable health coverage.
There are some challenges in identifying and quantifying job-lock due to the enactment of the individual mandate and the employer mandate, which corresponds to the same year as the Medicaid expansion and the creation of the Marketplace. The individual mandate is a tax penalty on non-poor individuals for not enrolling in health insurance. This can either incentivize individuals to supply their labor to firms offering ESI or to reduce working hours to keep income below the tax filing threshold ($20,800 for a married couple). The employer mandate is a penalty imposed on applicable large employers (having at least 50 full-time equivalent employees) who do not offer health insurance. The penalty is applied if a full-time employee purchases a plan in the private market and receive federal premium subsidies. This penalty could incentivize large firms to increase demand for part-time workers. In fact, William E. Even and David A. Macpherson show a substantial amount of increase in involuntary part-time employment resulting from the mandate.
Both mandates are less likely to affect the employment outcomes of Medicaid-eligible individuals, and hence, most studies prefer to use Medicaid beneficiaries to analyze job-lock under the ACA. On the other hand, any reduction in employment for non-Medicaid population could be a combined effect of the Marketplace and the mandates, leaving the job-lock effect ambiguous. Overall, it is challenging to make assertions about the impact of the ACA on job-lock, let alone the aggregate labor market effects, due to the different incentives created under each provision.
Based on data from the 2018 March Current Population Survey (CPS).
Note that people with incomes below 138% of the FPL in non-expansion states and those with incomes below the tax filing threshold are exempt from the individual mandate.