The IRS and HHS have been issuing guidance and notices, as well as regulations, regarding the definition of affordable coverage. But employers seem to be caught up in household income, 8%, 9.5% and what constitutes affordable coverage to avoid penalties. So breaking down this one issue of "affordability," let's summarize where we presently appear to be (at least as I understand the rules):
First, assume you are offering coverage that meets the minimum essential value test (which is still being defined). If the employee cost of obtaining "single coverage" is less than 9.5% of that employee's total compensation as reported in box 1 of that employee's W-2, that coverage will be "affordable." Coverage generally has to be available to cover dependents, excluding spouses, but the cost of dependent coverage is not calculated in the determination of whether the employer is offering affordable coverage. It may have an impact on the eligibility for subsidies of the employee, but not the employer. So, as it presently sits, if the cost of single coverage to the employee is less than 9.5% of their compensation, it is affordable.
Now, from the employee's perspective, whether or not the coverage offered by the employer is affordable looks at total household income. For the individual, their personal definition of whether they are being offered affordable coverage will be based on 8% of total household income. By way of example, think of an employee making $30,000 a year. The employer has to offer that employee single coverage for an employee contribution of $237.50 a month or less ($30,000 x 9.5% = $2850/12). The employer would be OK because that coverage would be affordable.
Now suppose that employee has two children. The employer can charge more for the family coverage than the 9.5% because that limit applies only to the single coverage. So let's assume the employer charges $500 a month to the employee for family coverage. The coverage would still be affordable and the employee would not be subsidy eligible if they went to the exchange because the employer is offering affordable coverage. The exchange coverage for the family might be a cheaper option and the employee could choose it, but they would not get subsidies.
Where things get interesting is when you consider non-covered spouse. Assume the spouse is unemployed and the total household income is that same $30,000. If the spouse goes to the exchange for coverage themselves, the spouse would be eligible for a subsidy because they are not offered any employer sponsored coverage and the total household income for the family is below 400% of the federal poverty level. Further, because of the household income is so low, it is likely that children could end up with coverage through a state-sponsored S-CHIP plan.
Now it gets even more confusing. If the employee decides to opt out of the employer coverage and the whole family remains uninsured, they would not be subject to the individual tax penalty for not having insurance because the cost of the employer coverage for the family is more than 8% of total household income ($2,400 versus $6,000). But that total household income calculation applies to the employee and family, not the employer. Back to step one, if the employer's single only coverage is less than 9.5% of the employee's income, it is affordable and the employer pay no penalty,
As you can see, this means that it would be hard to argue that this is going to make coverage more affordable for families. And employers that were charging employees more than 9.5% of the cost of coverage (or not offering coverage at all) will see some additional cost. But for employers, the point is you don't have to get caught up in trying to figure out the total household income of your employees. Look only at their individual income and base your affordability measure on that figure (at least for now).