Pay or Play Employer Mandate: Cost & Efficiency - Coverage Expansion

Model Summary

This type of proposal would require California employers and employees to pay a fee, based upon a percentage of payroll, to a state insurance fund for eligible employees (the "pay" option). Or employers offering a minimum level of health coverage would receive a credit against the fee (the "play" option).

For details and analysis on any of these attributes for this model, choose an item from the drop-down list.

Pay or Play Employer Mandate: Cost & Efficiency

Resource/Real Cost

The resource costs would be those associated with expansion of employer-based coverage under the play option and the expansion of state-provided coverage under the pay option. In other words, the real resource costs would equal the value of the additional medical resources the newly covered people would consume. Since the expansion of coverage would be only to employed people and their dependents not already covered, the net resource cost would be moderate.

Cost Containment

This approach includes no explicit cost control policies. Without new controls, costs would be expected to rise, as they have in the past, since the underlying factors that account for such increases would remain in place. Although the amount paid by employers that choose the "pay" option would be a fixed percent of payroll (at least as proposed here) and would thus rise as the economy and payrolls grow, the rate of tax revenue growth is likely to lag behind the rate of increase in health costs; over time, this differential would likely result in a shortfall in the state fund that finances the program for employees (and dependents) not offered coverage by their employers. Typical employers who chose the "play" option would likely find that their insurance costs rise more rapidly than their revenue. Because all employers are required to pay for coverage in one way or another and cannot drop coverage or reduce benefits below the standard package no matter how much costs rise, they could become a major source of support for efforts to contain costs if they rise significantly over time. (For a general discussion of tools for cost containment, see the Related Link below.)

Budgetary Cost

Because the costs of coverage for many newly covered people would be paid by those employers that decide to "play," this aspect of coverage expansion would be achieved without any additional budgetary cost to the state. The state would incur new costs to provide coverage for people whose employers decide to "pay," but if the fee level is properly set, most of the cost would be offset by the payroll assessment on employers and employees. Whether the fees would be adequate to cover costs would depend, first, on whether the assessment rate was adequate to pay for the benefits available under the state plan for enrollees of average risk and, second, the extent to which the state plan experienced adverse selection.

Probably most lower-wage employers would choose the pay option because if they bought coverage on their own, the cost would in most cases represent a higher proportion of payroll than the fee does (a maximum of 9% of payroll). So the adverse selection problem would probably not be severe for this group, but the revenues would almost certainly fall short of covering the cost of the medical claims incurred. The state would have to make up the difference from other sources. Setting the fee at a higher percent of payroll would reduce the need for other revenue for two reasons: fewer employers would choose the pay option and each participating firm would pay more. But adverse selection would probably increase.

Many firms whose employees are older or are of above-average risk would also likely chose to "pay" rather than "play,” because it would often be cheaper for them to pay the fee than to buy coverage in the regular insurance market or to self-insure. This adverse selection problem should be lessened by the feature of this prototype that extends small-group market rules to employers that employ 200 or fewer workers: because insurers' ability to risk rate these employers would be limited, higher-risk firms would be able to buy coverage in the private market on an adjusted community-rate basis, which would make them less likely to choose the pay option. However, some low-risk companies in the approximate range of 100 to 200 employees might decide to self-insure rather than buy insurance, so they would not be included in the insurance pool, which would drive up costs for others. If the state-provided pool experienced some adverse selection, the state would have to subsidize the cost of the adverse selection with new dollars from some other revenue source, presumably general revenues.

(Since this pay-or-play prototype was first published in 2004, the Institute for Health Policy Solutions produced a study that suggested that our original payroll fee percentages would be inadequate. In December 2005, we amended the prototype to increase the maximum employer payroll fee from 7.5% to 9% and the employee payroll fee from 1.5% to 2.5%. For the complete study, see "Challenges and Alternatives for Pay or Play Program Design — Supplement A.")

One advantage of payroll-based financing is that budgetary revenues would grow as the state economy grows, which is important because medical costs are certain to grow over time. Unfortunately, medical costs have consistently grown more rapidly than wages, which means that over time either the fee would have to be increased or the benefits offered to employees in the state plan would have to be reduced.

This model could result in a substantial loss in personal income tax revenue for the state. Economists generally agree that most of the employer's cost of newly extending health coverage to workers and/or dependents would be passed back to workers in reduced wages, perhaps not immediately but over time. That is, the value of total compensation would stay approximately the same, but employee take-home pay would be lower by any amount the company pays for new coverage under the play option or for the fee under the pay option. The reduction in take-home pay would be reflected in lower personal income tax collections because what the employer would pay for premiums or for the health coverage fee would not be taxable and would not generate revenue for the state.

The total effect of the state personal income tax reduction could be significant because the fee for the pay option would be 9% of total payroll, and the cost for employers who choose the play option would be 80% of the premium for employees and 70% for dependents. Seventy percent of a $8,000 to $10,000 family premium is a substantial sum and a significant percentage of current wages for lower-wage workers. The proportion of employers that now pay less than 70% for dependent coverage is relatively high, so the effect would apply to many employers who already offer coverage. In addition, for employers who chose the pay option, employees would pay a fee equal to 2.5% of wages, which would reduce their taxable income by that percentage.

Ease of Initial Implementation

For the state, initial implementation would require making a series of relatively complex administrative decisions, especially around setting up the state fund. No significant new bureaucracy should be necessary, and the task should be somewhat easier than it might otherwise be because the state's Managed Risk Medical Insurance Board already administers a number of somewhat similar programs and could take on this task as well. Employers who chose the pay option should experience minor administrative costs: they would simply pay the state a fee that is a percent of payroll.

To make the transition from the status quo easier for medium-sized and smaller employers, the fee would be phased in over a series of years, starting with a smaller percentage of payroll and gradually increasing. Employers choosing to newly cover their employees would experience the administrative costs of arranging for coverage, determining eligibility, paying premiums, etc.—all of the administration associated with offering health coverage. However, any employers who found this too burdensome could opt instead just to pay the fee. Overall, there would be little disruption of the status quo and only modest need for new regulatory machinery.

Ongoing Administration/Regulation

Once this program was underway and the initial start-up costs incurred, it should not be expensive to administer either for employers or the state.

Legal and Regulatory Issues

The Employee Retirement Income Security Act (ERISA) contains a broad provision designed to preempt all state laws that attempt to regulate employee benefit plans sponsored by private-sector employers or unions. As a result, ERISA prevents states from directly regulating health coverage arrangements established by employers but allows states to regulate the indemnity insurers and health plans with which employers contract. This pay or play employer mandate may be vulnerable to a legal challenge under ERISA because it conditions a waiver of the fee on an employer’s health coverage plan meeting a specific benefits definition and covering a minimum 70% premium share. These criteria for the waiver may cause preemption because they can be argued to "relate to" employer-sponsored health plans by virtue of affecting the plans’ benefits, structure, and/or administration—the standard set out by the Supreme Court in thirty years of preemption cases. Supporters of the law could argue that ERISA should not preempt the law because claiming the credit is voluntary. However, if an employer did not want to claim the credit (for example, in order to offer a standard set of benefits or premium contributions across multiple states that differed from the required benefits and premium contribution under this mandate), paying the fee while also providing coverage could impose a substantial economic burden that might cause ERISA preemption (based on language in the Supreme Court’s Travelers opinion regarding onerous burdens that has not yet been applied by any court). No courts have addressed precisely this type of state-based public program for employees funded partially by employers with a waiver or credit conditioned on the employer’s plan containing certain features, but it might be difficult for one with this set of features to overcome a preemption challenge.

Some legal experts think it is possible to design a pay or play law to avoid ERISA preemption problems by offering a dollar-for-dollar credit for the cost of any employee health benefits offered, without regard to the "adequacy" of the benefits, premium contributions, or other features.

If the program funding is characterized as a "fee" rather than a "tax," it also could be challenged under California Constitution Section 13A.

See the Related Links below for a paper that briefly outlines why ERISA raises problems for this type of state health policy initiative and how recent Supreme Court decisions have reduced ERISA’s preemptive impacts.

Labor Force Effects

Employers who do not offer coverage would experience an initial labor cost increase under either the pay or play options, although no employer would need to pay an amount more than 9% of payroll, since employers could choose the pay option, limiting the cost to 9%. However, as noted, economists generally believe that in the longer run these costs would be passed back to employees in the form of reduced wages or cuts in other benefits. Total compensation costs for most employers would, thus, be little affected in the long run. Most employers, even in the short run, would not be likely to lay off significant numbers of workers as a result of the change in labor costs.

This might not be true, however, for employers paying at or near the minimum wage. Existing minimum wage laws make it impossible for minimum-wage employers to pass back the costs to employees in the form of reduced money wages. The increased labor costs might force some employers to hire fewer workers or lay off some of the employees that produce the least revenue for the firm per hour of work—typically low-wage workers such as teenagers. If over time the costs of coverage increased more rapidly than worker productivity (as has typically been the case), this effect could persist over a number of years.

To put this in perspective, the labor force effect should be similar to a 9% increase in the minimum wage phased in over several years—five years for medium-sized employers and seven years for small employers. Past experience suggests that no major labor disruptions would be likely. The 9% fee would apply to all payroll, so the financial impact would be greater than a 9% increase in the minimum wage, but the fee increase would probably be passed back to the higher-paid workers in reduced wages or benefits in the long run and would thus not have a major impact on the number of people hired. The fact that the fee percentage rises gradually for all but the largest employers would make it easier for employers to pass the cost back to employees even in the short run. Of course, employers who decided for the first time to "play" and offer coverage would not benefit from the phase-in, but presumably they would choose that option rather than the pay option only if they decided that it was to their advantage to do so.

The short-run effect on employment would also depend on what proportion of a firm's total costs are labor costs. If labor costs are a relatively small portion, it would be easier for the employer to absorb or pass on to their customers the amount the employer paid for the new fee.

Employers would have an incentive to shift people from full-time to part-time work to avoid covering them or paying the fee, but their ability to do so would be limited by the fact that if an employee has accumulated 300 hours of work in the last 12 months, they must be covered. Moreover, shifting employees from full-time work to part-time work would be illegal under laws and regulations that would be implemented with this model. Still, violations could be difficult to prove, so that the prohibition could be difficult to enforce.

Accountability for Ensuring Efficiency and Quality

For the most part, this approach leaves accountability for efficiency and quality essentially as it is with the status quo. Employers could still seek to promote efficiency and quality by working with health plans to promote these ends, as many do now. The state pool could use its leverage with participating health plans to achieve these ends as well. But no great changes from the status quo would be expected.