Perspective
“Play-or-Pay” Insurance Reforms for Employers — Confusion and Inequity
N Engl J Med 2010; 362:93-95January 14, 2010
- Article
One prominent feature of the current health care reform bills is a “play-or-pay” rule for employers: workers must receive part of their compensation in the form of employer-sponsored health insurance, or the company's payroll will be subjected to a tax penalty. Although the latter possibility is framed as a penalty on employers, it would almost surely be passed along as a levy on workers' wages.1 And although the “play” option is usually described as involving employer “contributions” toward insurance, this money, too, actually comes out of what would otherwise have been workers' wages.2,3 Rather than fostering transparency, the play-or-pay approach results in confusion about who is paying how much for what coverage — and careful analysis indicates that this method would not address existing inequities and could add further inequities to our system.
Consider an insurance policy designed to meet the standards outlined by the proposed new insurance exchange. If an employer chose the “play” option, the premium for that policy would be shielded from income and payroll taxes, as it is today. If the employer chose the “pay” option, it would have to pay a tax penalty, but its workers could buy insurance through the exchange, with a subsidy that decreased as income increased. Workers would prefer the arrangement offering the larger net subsidy. The subsidy in the “play” option would continue to be the worker's combined marginal tax rate times the standard premium (the taxes the employee would have to pay if there were no tax exemption for this form of compensation); the magnitude of this subsidy increases as wages increase. The net subsidy under the “pay” option would be the explicit government subsidy offered for purchasing insurance through the exchange, minus the tax penalty passed along by the employer; the magnitude of this subsidy would decrease as wages increased.
As a result, workers would tend to prefer “play” over “pay” as their wages increased. If every company paid all its workers the same wages, there would be some wage level below which all workers would want their employers to “pay” and above which they would want their employers to “play.” The result is a V-shaped pattern: subsidies would decrease with wages for those purchasing insurance through the exchange and then increase with wages for those with employment-based insurance. This pattern certainly doesn't match the goal of vertical equity.
In reality, of course, most companies employ some mix of low-wage and high-wage workers, though some have lower-than-average wages overall and others have higher-than-average wages. We expect that low-wage companies would tend to choose to pay a penalty, and high-wage companies to provide insurance. (Although the penalty is described as an explicit employer-paid tax on a company's total payroll, it would be an implicit employee-paid “tax” on workers' wages; the portion of the tax passed on to each worker as a wage reduction would be the tax rate times that worker's wage, since higher-wage workers would have a greater impact on the total assessment.)
The graph
Net Subsidies under the House's Health Care Reform Bill. shows the pattern of subsidies under the Affordable Health Care for America Act passed by the House of Representatives in November. We consider a hypothetical low-wage company that chooses the “pay” option and is subject to the 2% payroll assessment applicable to companies with a payroll of $500,000 to $585,000 (e.g., a 20-person company with average wages of $27,500). An 8% assessment would be applied to companies with a payroll of more than $750,000. The “pay” option's pattern of subsidies, which decrease with income, would be similar under the Senate's Patient Protection and Affordable Care Act, although the Senate bill uses a $750 penalty per worker on companies with more than 50 workers, rather than a percentage of total payroll. We also consider a hypothetical high-wage company that chooses the “play” option.Low-wage workers in a high-wage company would be worse off than low-wage workers with identical productivity in a low-wage company. For instance, a single worker earning $21,660 — 200% of the federal poverty level for an individual — would receive a net subsidy of $3,574 through the exchange if he or she were employed at a low-wage company choosing to “pay” but would get a subsidy (a tax exemption) of only $1,887 if employed at a high-wage company choosing to “play.” The $1,687 difference represents about 32% of the premium and 8% of the worker's income. For low-wage workers who are currently uninsured, such a difference might have a substantial effect on compliance with a mandate to obtain insurance.
There is also horizontal inequity among high-wage workers. For instance, a worker earning $43,320 — 400% of the federal poverty level for an individual — would have to pay $866 (2% of payroll) in lower wages if he or she were employed at a low-wage company that opted to pay a tax penalty but would effectively receive a subsidy of $2,407 if he or she were employed at a high-wage company that opted to provide insurance. The difference is about 63% of the premium and 8% of income. Moreover, although the House bill specifies that premiums for people at 400% of the poverty level would be capped at 12% of income, they would actually amount to 14% after the 2% payroll assessment was added. And if workers with incomes at 400% of the poverty level were employed by a larger low-wage company that had to pay the full 8% assessment, they would pay $3,466 (8% of payroll) in lower wages, for a total contribution of 20% of their income (this 8% plus a 12% premium). So high-wage workers would be worse off in low-wage companies than in high-wage companies.
Thus, because each company's decision to play or pay would be driven by its average wages, heterogeneity within companies would cause the subsidies for many individual workers to be mismatched with their level of need. Moreover, tax penalties and subsidies that depend on a company's size would result in further inequity among low-wage companies, because subsidies for workers with the same income would be larger if they worked for small companies (which had to pay smaller penalties) than if they worked for large ones. Policymakers expressing a desire to help “small business” seem to ignore the fact that benefits go to people, not companies, and that variations in payroll taxes primarily affect workers, not business owners.
We believe there is a way to avoid this confusion and inequity. Ideally, the subsidy for private insurance should depend explicitly on total family income. Such a program would be described as one in which workers (not employers) pay for coverage, but the employer might arrange for group insurance and collect wage-related taxes and premiums.4 The current inequitable tax subsidy would be replaced, and the concept of a penalty for companies not offering coverage would be abandoned. The proposed subsidy would instead be inversely related to income, and people would receive the same size subsidy for the same coverage regardless of whether they obtained their insurance through an exchange or their employer. For people with employment-arranged insurance, the premium's value would become taxable income, but the additional tax cost borne by workers would be offset by a progressive income-related subsidy toward the premium, administered as a tax credit either directly to employees or indirectly through employers and exchanges.
If such a system seems too radical, however, two relatively modest changes could be made to the legislation under consideration: we could widen the scope of subsidies for low-income Americans in order to improve horizontal equity, making more low-wage workers eligible for the same subsidy regardless of where they work; and we could cap the tax exclusion for employment-based insurance, perhaps for policies with high actuarial value, for higher-income workers, or both. The resulting tax revenue could be used to help fund the more equitable low-income subsidies. These steps would at least begin to move us in the direction of improved equity.
Financial and other disclosures provided by the authors are available with the full text of this article at NEJM.org.
This article (10.1056/NEJMp0911920) was published on December 30, 2009, at NEJM.org.
Source Information
From the Department of Health Policy and Management, Bloomberg School of Public Health, Johns Hopkins University, Baltimore (B.H.); and the Health Care Systems Department, Wharton School, University of Pennsylvania, Philadelphia (M.V.P.).
- References
References
1
Historical effective federal tax rates: 1979 to 2006. Washington, DC: Congressional Budget Office, April 2009.
2
Pauly MV. Health benefits at work. Ann Arbor: University of Michigan Press, 1997.
3
Gruber J. Health insurance and the labor market. In: Cuyler AJ, Newhouse JP, eds. Handbook of health economics. Amsterdam: Elsevier, 2000:645-700.
4
Pauly MV . Making a case for employer-enforced individual mandates. Health Aff (Millwood) 1994;13:21-33
CrossRef | Web of Science | Medline
