This column is a collaboration between KHN and The New Republic.
McKinsey and Company has finally released the details of its controversial paper on the likely effects of health care reform. And it looks like the paper’s critics (including yours truly) were right to raise questions about it. Based on what the company has said, the paper offers no new reason to think Americans with employer-sponsored insurance will lose that coverage because of the Affordable Care Act.
Politically, that’s good news for President Barack Obama, since he told insured Americans that the law wouldn’t take away the coverage they already had. But what does it mean in terms of policy? Should we be happy that health care reform is unlikely to reduce substantially the current system’s dependence on employer-based insurance? That’s another, much more complicated question.
Let’s start by looking closely at the paper, which suggested that as many as half of all employers would seriously readjust their employee benefits and as many as a third would drop coverage altogether once the health law took full effect in 2014. The theory behind the claim was that employers would find it financially advantageous to stop offering insurance because workers could instead get subsidized coverage through the new insurance exchanges. In other words, workers would happily take salaries — instead of insurance — from their employers. Even though firms that employ more than 50 employees — which account the majority of American jobs — would have to pay a penalty for failing to offer health benefits, the McKinsey consultants said, the financial advantages of dropping coverage would more than offset that cost.
But the basis for that prediction was a survey of high-ranking corporate officials. And it turns out the survey had a few weaknesses. For one thing, a quarter of respondents didn’t know the salary breakdowns of their companies — in other words, how many workers were making high salaries and how many were making low salaries. In addition, more than half of respondents weren’t even aware of what their companies spent on health benefits. The survey didn’t ask respondents about the ages of their employees. Were they relatively old? Young? A mix of the two? And when survey administrators “educated” respondents about the health law, they didn’t remind them about the effects of the employer tax exclusion, which makes job-based health insurance worth a lot more to employees.
Surveys asking employers to predict behavior are never that reliable. But these issues make the McKinsey study a particularly poor forecasting tool. In general, younger and poorer workers might be better off getting insurance through the new exchanges, because they will get bigger subsidies from the government and because they benefit less from the tax exclusion. It is a different story for older and richer workers.
More sophisticated studies of employer behavior account for these and other variables, typically by creating “synthetic” firms and predicting how employers will act, based on data on past employer behavior. Sure enough, these studies have consistently shown a very different result: that the majority of employers will continue to offer health insurance, even after health care reform. In fact, just this week, new analyses by Avalere Consulting and the Robert Wood Johnson Foundation came to the same conclusion. (The basis for the Robert Wood Johnson prediction was another projection from the Urban Institute’s model.)
While these predictions could be wrong, obviously, their findings are consistent with what happened in Massachusetts, where a similar coverage scheme actually bolstered employer-sponsored insurance. Indeed, even McKinsey itself now acknowledges that its study couldn’t make projections as reliably as these other efforts. In its official June 20 press release, it stated flatly that its prediction was “not comparable to the health care research and analysis conducted by others such as the Congressional Budget Office, RAND and the Urban Institute.”
But here’s the irony: Most people like the insurance they get from their employers, which is why you hear politicians from both parties constantly promising to keep that coverage in place. In the long run, though, workplace-based insurance is probably not an arrangement worth preserving.
Private, employer-based coverage became the norm in the U.S. in the 1940s and 1950s because the arithmetic of health insurance works only with large groups of relatively randomly selected people, and large businesses naturally create such groups. But employer-based coverage makes workers too dependent on their bosses while saddling employers with a financial liability over which they have only partial control. More importantly, it leaves out people who don’t have steady, full-time work.
An ideal health care system would not merely include everybody. It would also give everybody access to the same set of coverage arrangements, regardless of their place of employment (or lack thereof). It would also liberate employers from the responsibility of administering health benefits for workers, allowing them to concentrate on other, more productive activities. Let the car companies make cars and the grocery stores sell groceries and the software firms design software. They don’t need to be running health insurance plans, too.
A single-payer system, with a combination of basic government insurance and private supplemental coverage, would be a much better alternative. So would a “competition” system that looks like what is currently in place in the Netherlands or Switzerland, or what Sen. Ron Wyden, D-Ore., first proposed back in 2007. The Affordable Care Act could evolve into such a system, particularly if the new insurance exchanges work well and workers feel comfortable the insurance available there is as good as what they’d get from employers. But that transition would probably take a lot of time, no matter what corporate officials were telling the survey-takers at McKinsey.
Jonathan Cohn is a senior editor at The New Republic .
Some elements may be removed from this article due to republishing restrictions. If you have questions about available photos or other content, please contact firstname.lastname@example.org.