Health economist John C. Goodman once said, at a talk I attended when Obamacare was being debated, “Would you want to be a patron of a restaurant that didn’t want you as a customer?”

He was getting at the fact that Obamacare would set up incentives for health insurers to avoid the sickest people because they would not be able to charge for pre-existing conditions. Under the individual insurance that existed before Obamacare, beneficiaries (my daughter was one) could buy guaranteed-renewable health insurance. If they developed a condition while insured, they could still buy health insurance at a premium that applied to the whole pool they were in when they originally bought insurance. If they developed a serous condition, why would insurers take them? Because, by contract, they had to. Why would they treat them well? Because they had an incentive to treat everyone in the pool well and could be caught for discriminating against people in the pool.

Obamacare looked good because insurers could no longer charge for pre-existing conditions. What to do? According to a study by Michael Geruso of the University of Texas, Timothy J. Layton of Harvard Medical School, and Daniel Prinz of Harvard University, health insurers under Obamacare have figured out how to game the system: how, in Goodman’s terms, to be the restaurant that turns away customers they don’t want. Here’s a segment of their abstract:

We first show that despite large regulatory transfers that neutralize selection incentives for most consumer types, some consumers are unprofitable in a way that is predictable by their prescription drug demand. Then, using a difference-in-differences strategy that compares Exchange formularies where these selection incentives exist to employer plan formularies where they do not, we show that Exchange insurers design formularies as screening devices that are differentially unattractive to unprofitable consumer types. This results in inefficiently low levels of coverage for the corresponding drugs in equilibrium.

The study is “Screening in Contract Design: Evidence from the ACA Health Insurance Exchanges,” NBER Working Paper #22832, November 2016.

Cato Institute health economist Michael Cannon lays out some of the implications in a recent op/ed titled “How ObamaCare Punishes the Sick,” Wall Street Journal, February 28, 2017 (March 1 for print edition.)

A long excerpt:

Predictably, that triggers a race to the bottom. Each year, whichever insurer offers the best MS coverage attracts the most MS patients and racks up the most losses. Insurers that offer high-quality coverage either leave the market, as many have, or slash their coverage. Let’s call those losses what they are: penalties for offering high-quality coverage.

The result is lower-quality coverage–for MS, rheumatoid arthritis, infertility and other expensive conditions. The researchers find these patients face higher cost-sharing (even for inexpensive drugs), more prior-authorization requirements, more mandatory substitutions, and often no coverage for the drugs they need, so that consumers “cannot be adequately insured.”

The study also corroborates reports that these rules are subjecting patients to higher deductibles and cost-sharing across the board, narrow networks that exclude leading cancer centers, inaccurate provider directories, and opaque cost-sharing. A coalition of 150 patient groups complains this government-fostered race to the bottom “completely undermines the goal of the ACA.”

It doesn’t have to be like this. Employer plans offer drug coverage more comprehensive and sustainable than ObamaCare. The pre-2014 individual market made comprehensive coverage even more secure: High-cost patients were less likely to lose coverage than similar enrollees in employer plans. The individual market created innovative products like “pre-existing conditions insurance” that–for one-fifth the cost of health insurance–gave the uninsured the right to enroll in coverage at healthy-person premiums if they developed expensive conditions.

There are lots of links in the above, so, if you want to know more about what’s behind his reasoning, do a Google on the title of his article.

He also warns policy makers who are hesitant about repealing the pre-existing condition rules:

If anything, Republicans should fear not repealing ObamaCare’s pre-existing-conditions rules. The Congressional Budget Office predicts a partial repeal would wipe out the individual market and cause nine million to lose coverage unnecessarily. And contrary to conventional wisdom, the consequences of those rules are wildly unpopular. In a new Cato Institute/YouGov poll, 63% of respondents initially supported ObamaCare’s pre-existing-condition rules. That dropped to 31%–with 60% opposition–when they were told of the impact on quality.