As it enters its fifth year, the Affordable Care Act has chalked up an impressive list of accomplishments. More than 8 million Americans have chosen a health plan through the ACA exchanges.  At least another five million have likely enrolled in MedicaidThe minimum medical loss ratio requirement has saved privately insured Americans billions of dollars, while the closing of the doughnut hole has saved Medicare beneficiaries billions more.  The percentage of Americans who are uninsured is dropping precipitously and is already at the lowest level it has been for years.

Recent polling, however, seems to show that Americans are not yet impressed —  a majority still oppose the Affordable Care Act.  Significantly, however, polling also consistently shows that Americans are not giving up on the law–a substantial majority of Americans are against repealing the ACA and would rather that problems with the ACA be fixed.  Support for amending the law should create an opening for lawmakers who can identify real problems with the ACA and propose practical solutions.

One provision of the ACA that cries out for repair is the employer mandate. The Urban Institute has recently raised the question, “Why Not Just Eliminate the Employer Mandate?”  Conservative advocacy groups have called for its repeal for some time.   Repeal of the employer mandate might, in fact, not be such a bad idea, as long as the current mandate was replaced with a better alternative.

The employer mandate.  The ACA really contains two “employer responsibility requirements.”  One requires large employers (employers that employ fifty or more full-time or full-time equivalent employees) to offer “minimum essential coverage” to all of their full-time employees or pay a $2,000 penalty for each and every full-time employee if any one employee receives premium tax credits through the exchanges.  A second requirement imposes a $3,000 penalty on a large employer for each employee who receives premium tax credits for exchange coverage if the employer fails to offer that employee affordable and adequate coverage.

It is clear why the ACA includes an employer mandate.  Fifty-five percent of Americans are currently insured through their place of employment, including over 90 percent of those for whom private insurance is their primary form of insurance.  Congress was concerned that the ACA’s initiatives to expand coverage of the uninsured by fixing the dysfunctional individual market might undermine a basically well-functioning group insurance market.  Congress enacted the employer mandate to ensure that employers would continue to offer coverage to their employees and not dump them into the individual market, where their coverage would be paid for in part with federal tax credits.

Employer mandate problems. The employer mandate, however, is resulting in a host of problems. To begin, implementing the mandate has proven to be an incredibly complex task. As already noted, the mandate only applies to large employers, those with at least 50 full-time or full-time equivalent employees.  But an employer’s number of employees is not always easy to count, and the 50-employee threshold seems quite arbitrary.

Second, employers are only required to insure full-time employees.  The ACA defines a full-time employee as one who works at least 30 hours a week. But workers’ hours often fluctuate.  And while counting hours can be fairly straightforward for hourly wage workers, it is more difficult for salaried workers.  For many employees — adjunct faculty, seasonal farm workers, airline pilots, or high school coaches, among others — counting hours makes little sense at all.

The ACA allows employers to impose a waiting period of up to 90 days before covering workers.  But determining when the 90 days begins to run is sometimes itself complicated — for example, when an employee moves from part-time to full-time status.  To make the law work, employers must report to the federal government and to their employees a great deal of information regarding the health benefits they offer and to whom they are offered.  Employers will no doubt spend hundreds of thousands of hours compiling this information and the government thousands more hours processing it.

The employer mandate is also already having an effect on business arrangements and on the labor market, although the full extent of this effect is difficult to know.  Small employers approaching the 50-employee threshold have reportedly constrained growth to avoid becoming subject to the mandate.  Some employers have cut the hours of part-time employees to below 30 to avoid having to offer them coverage.  Other employers have reportedly dropped coverage for spouses of employees who had access to other coverage, as the rules implementing the mandate do not require spousal coverage.

Despite all of this, it is not even clear that the employer mandate will actually result in decent coverage for employees.  To avoid the $2,000-for-every-full-time-employee penalty, employers need simply offer “minimum essential coverage.”  “Barebones” coverage, however, which pays for preventive services and does not violate other specific provisions of the law, qualifies as minimum essential coverage.  Employees who accept “barebones” minimum essential coverage are exempt from the individual mandate penalty and thus need not obtain real insurance, even though they are effectively uninsured.

If an employer additionally offers adequate coverage (with at least 60 percent actuarial value) that is affordable (can be purchased for 9.5 percent or less of an employee’s wages), the employer escapes all penalties, even if not a single employee accepts coverage. If an employer fails to offer coverage to a low-income employee, moreover, and that employee ends up on Medicaid, the employer suffers no consequences.

It gets worse. Because the ACA is built on the assumption that individuals with employee benefits have coverage, it prevents individuals who accept employer coverage — however barebones and inadequate — from qualifying for premium tax credits and cost-sharing reduction payments.  Even an offer of “adequate” and “affordable” coverage can disqualify individuals from tax credits, whether or not it is accepted.  Indeed, the offer of affordable individual employer-sponsored coverage to an employee disqualifies the employee’s entire family from subsidized exchange coverage, even if family coverage is clearly unaffordable.

Individuals are required by the individual mandate to purchase health insurance, but if employer coverage is unaffordable (costs more than 8 percent of household income), they are exempted from the requirement. But the tasks of determining whether an individual should be exempt from the mandate because employer-sponsored insurance is not affordable, or whether an individual is eligible for premium tax credits because employer-sponsored coverage is unaffordable or inadequate, add greatly to the work of the exchanges.

The Administration’s response.  Not surprisingly, the administration has delayed enforcement of the employer mandate as it struggles with its complexities.  Unable to figure out how to implement the employer reporting requirement, it announced in the summer of 2013 a delay of both the employer reporting requirement and the mandate itself until 2015The administration subsequently announced that employers would only need to achieve 70 percent compliance for 2015, while also delaying the mandate for employers with between 51 and 100 employees and required coverage for dependents until 2016 under certain circumstances.  Even after 2016, the final regulations implementing the mandate require only 95 percent, rather than full compliance.

Rules defining minimum coverage are not yet final, and the IRS has indefinitely postponed the ACA’s automatic enrollment and non-discrimination-in-favor-of-highly compensated employee requirements.  Another requirement — that employers designate on W-2s the cost of health benefits — is only partially implemented.  The administration has recently announced that state exchanges have until 2016 to fully verify fully employer coverage for premium tax credit applicants.

Why not just repeal it?  If the employer mandate is so problematic, why not simply repeal it?  One obvious argument against repeal is that without the mandate employers will drop coverage, leaving their employees uninsured or forced to seek tax-subsidized coverage through the exchanges. This will impose greater costs on the federal government and on employees themselves.

There are good reasons to believe, however, that most employers would continue to provide coverage in the absence of the mandate.  The vast majority of employers subject to the mandate offered coverage before it went into effect (99 percent of employers with more than 200 employees and 91 percent of employers with between 50 and 199 employees).  Health benefits have long been regarded as essential for recruiting and retaining skilled and competent employees, and they contribute to increased productivity and reduced absenteeism.  Employer coverage can usually be provided with lower administrative expenses than individual coverage, and thus provides good value for employers and employees.

Perhaps most importantly, employment-based benefits are already heavily subsidized through the tax system.  They are exempt for both the employer and employee not only from the federal income tax, but also from Social Security, Medicare, and federal unemployment tax, as well as state income tax.  The employer mandate penalties impose only a marginal additional incentive to obtain coverage.

It is likely, however, that some employers currently providing employee benefits would drop benefits for some employees in the absence of the mandate, and that some employers who do not currently offer benefits may start to do so if the mandate remains in place.  The Urban Institute study concluded that the number of uninsured would only increase by 200,000 in the absence of the employer mandate.  Other analyses offer greater numbers—the Congressional Budget Office projects 500,000 to 1 million.   This is not a trivial concern.

Moreover, the repeal of the employer mandate could have other effects. First, it would increase the temptation for employers who currently offer coverage to most employees to drop coverage for low-income employees or employees with high-cost medical conditions and send them to the exchange.  There are already laws that discourage employers from doing this — prohibitions against health status underwriting on the one hand and rules limiting discrimination in favor of highly-compensated employees in self-insured plans, on the other — and the employer mandate itself contains loopholes that employers can exploit to engage in discriminatory conduct anyway.  But the incidence of employee dumping would likely increase if the mandate were repealed.

Many employees dropped from coverage would face increased costs, even though many of them would qualify for federal subsidies through the exchange.  Indeed, many Americans with employer coverage have incomes above 400 percent of poverty and thus do not qualify for exchange subsidies.  Even those who would qualify for subsidies might well pay more for coverage in the exchanges, and face higher cost sharing, than they would with employer coverage.

The end of the employer mandate, and the reporting requirements that accompany it, would also make the exchanges’ job of determining eligibility for premium tax credits and for exemptions from the individual mandate more difficult.  As already noted, eligibility for tax credits and for the individual mandate exemption turns on employee coverage offers and enrollment.  If employer reporting were eliminated together with the mandate, precise verification of whether an employee is eligible for coverage and the extent and cost of that coverage might not be possible.

Finally, and perhaps most importantly, repeal of the employer mandate would come at considerable cost to the treasury.  First, there is the obvious reduction in revenue the repeal would entail from lost employer mandate penalties. Second, the federal government would experience increased costs as individuals who would have been covered through their employers become eligible for premium tax credits and cost-sharing reduction payments or Medicaid.  These costs would be offset to some extent by the increased tax revenues the federal government would receive as employers and employees lose the tax subsidies that would have accompanied employer coverage.  But in the end, the CBO has estimated, the loss of the employer mandate would cost the federal treasury $130 billion over ten years (2014 to 2023).  The Urban Institute estimated the number would be closer to $46 billion, but the Rand Corporation said the cost could be $149 billion.

Is there a better way?  Is there a way to address these issues while avoiding the problems created by the current mandate?  Certainly.  Congress could look to earlier reform proposals, such as that adopted by Maryland in 2006 (and later invalidated as preempted by the then federal law).  The law required very large employers to spend at least 8 percent of payroll on employee health insurance or pay the difference between what they actually spent and 8 percent of payroll as a tax.  Adopting this approach would dramatically reduce the complexity of the current approach.  Employers would only need to know two numbers: the amount of their payroll and the amount they spent on health benefits.  The required percentage could be reduced for smaller businesses.

This proposal not only reduced the number of the uninsured, it was projected by the CBO to raise $135 billion for the federal treasury, essentially the same amount as would be raised by the current mandate.  Adopting this approach would dramatically reduce the complexity of the current approach.  Employers would only need to know two numbers: the amount of their payroll and the amount they spent on health benefits.

Implementing this tax would, of course, pose its own definitional complexities.  Employer aggregation rules would need to be applied to keep employers from forming shell entities to reduce their liability, and expenditures on health benefits would have to be defined, excluding for example flexible spending contributions under salary reduction agreements.  The statute would need to clarify that all compensation of all common law employees must be counted as payroll to minimize evasive strategies.  But implementation and reporting requirements would be infinitely simpler than those required by the current mandate.

Of course, other problems would remain.  Employers not subject to a universal mandate could spend the entire 8 percent of payroll on higher-income employees, sending their low-income employees into the exchange, or cover their healthy employees but dump their unhealthy employees into the exchange.  The first issue could be addressed in part by requiring employers to spend some lower percentage of payroll, perhaps 6 percent, exclusively on coverage for employees who were not highly compensated (using current IRS definitions), or, if an employer preferred, to offer the same benefits to all full-time employees (as the employer reasonably defined full-time), and pay a flat 8 percent.  The problem of dumping unhealthy employees could be addressed by continued enforcement of rules that prohibit discrimination on the basis of health status and the enactment of an anti-dumping provision, like that found in the ACA for the preexisting condition high risk pool, that would penalize employers found guilty of dumping.

To facilitate exchange determinations of eligibility for tax credits and for the individual mandate exemption, employers could be required to disclose to employees an employee’s required contribution for coverage at the time of open enrollment, or the time an employee actually enrolled if another time, as well as with the W-2 at the time of tax filing.  This amount should be easily determinable, as it is already deducted from the employee’s paycheck if the employee accepts the offer of coverage.  The employer could also provide the employee with proof of coverage if the employee accepts coverage, both at the time the employee accepts coverage and on the W-2.

The employer could even provide the employee with a statement as to whether coverage provides minimum value, as is currently required on the summary of benefits and coverage. This information could be provided by the employee to the exchange, and would allow the exchanges to determine whether the employee had access to affordable coverage and was in fact insured at the time the employee applies for exchange coverage or seeks an exemption from the mandate penalty.

Alternatively, the ACA could be amended to simply allow employees to choose between exchange coverage and employer coverage. This could greatly simplify reporting and eligibility requirements and would allow low-income individuals access to the most affordable coverage available to them. Employers could even be permitted to purchase individual coverage for their employees with tax-advantaged dollars, a practice that is currently illegal.  Freed from its connection with the employer mandate, the individual responsibility requirement could be amended to require that individuals in fact have “minimum coverage,” which would cover at least 60 percent of their medical costs, rather than the largely illusory “minimum essential coverage,” which imposes no real minimum for employer coverage. Employers would face an incentive to offer affordable and adequate coverage to their employees because they still have to meet the percentage of payroll requirement.

Of course, this approach would lack some of the precision and comprehensiveness of the current requirements. Some employees who would receive coverage under the current rules would be reclassified as part-time and lose coverage under the amended requirement, although they could get exchange coverage. This approach would also disadvantage some employers who would come out better under the current rules than they would under the percentage of payroll rules. Whatever benefits would be lost under the percentage of payroll approach, however, would be more than compensated by the reduction in complexity and deadweight loss administrative costs. The percentage of payroll approach should also have far fewer unintended labor market effects.

Is such an amendment politically possible?  Could such an amendment be adopted by Congress?  Diehard opponents of the ACA and some business interests would no doubt hold out for total repeal rather than agree to an amendment that would improve the law. Opponents may also insist on a repeal of the individual mandate if the employer mandate is to be fixed.  The individual mandate is far more important than the employer mandate to expanding coverage under the ACA, and although aspects of it could be reconsidered, this must happen separately.

The American people want the ACA to be fixed. I believe the Obama administration also would support an effort to fix the mandate, as it has signaled by its repeated delay of the requirement’s implementation. Candidates running for the House and Senate this year should be actively seeking ways in which the law can be improved. They should consider the approach offered here as one that could provide widespread benefits and could garner widespread support.

Editor’s note:  An earlier version of this post incorrectly described the approach of the 2009 House health reform legislation to employer responsibilty.  This has been corrected.