In the early stages of Obamacare, one thing was crystal clear to all health plans, stakeholders, and the government – it’s going to be really hard to have a brand new setup that reevaluates how Americans purchase and consume health insurance. The task ahead for everyone was nothing short of herculean, and not everyone was prepared for it. As a result, 2013, the year of the rollout of Obamacare, saw nationwide failures and delays in the federal as well as state health insurance marketplaces. While struggled with lost enrollments and broken transactions, state marketplaces held out for a little longer.

In time, the federal marketplace rebounded by fixing its internal mechanics and becoming a solid entity, ultimately enrolling about 12 million by the end of recent open enrollment in Feb 2015. On the other hand, state marketplaces, which had a better start, had to bite the dust. 7 states have decided to shut their exchanges and default to the federally facilitated marketplace, primarily because they haven’t been able to operate their exchanges profitably.

So what went wrong with these states? Let’s start with Hawaii, which recently decided to drop its exchange and move to federal system. Hawaii bagged $205 million from HHS for establishing its own exchange, but was able to enroll only 37,000 people in its exchange till date. For being profitable, Hawaii had to enroll 70,000 people to sustain the exchange running costs. Since that did not happen, Hawaii squandered its federal grant in running the exchange, but failed to make it work. Now, due to its latest decision of moving to the federal exchange, it will incur a bill of another $30 million – which will be covered by the taxpayers.

For Hawaii, there were two primary culprits which led to its demise – poor estimation and a failing system. Hawaii’s exchange was established by CGI group, and like its other exchanges, CGI received a whole lot of flak for a poor system that just did not help in enrolling the people. The other reason behind Hawaii’s failure plagued other states as well – poor estimation and miscalculation of prospective enrollees. HHS and states made some poor estimations while calculating how many people would enroll through these exchanges, and this calculation figured in allocation and distribution of federal funds for running the exchanges.

While miscalculations and poor estimations definitely laid the foundation for failure, the major blame lied with contractors who failed in their task in each of these states. Oregon, for instance, received $305 million in federal grants for its Cover Oregon exchange. Oregon had Oracle running the show for their exchange, but their spectacular failure and repeated failed attempts to fix the Cover Oregon exchange made the state lose all of its money in the exchange. Like Hawaii, Vermont too received a generous funding of $200 million for running its exchange. Vermont Health Connect was managed by CGI as well, and we know what fate all CGI run exchanges went through. Maryland also wasted about $183 million in federal funds, while New Mexico’s exchange floundered $122 million on its exchange.

The case of Nevada and Massachusetts was a little different. Nevada had $100 million in grants, and it chose Xerox to establish Nevada Health Link. The Nevada Health Link performed poorly throughout, and within 8-9 months of operation, the board decided to drop Xerox as its provider in May 2014. Massachusetts, which accepted $176 million from HHS, chose CGI to design and run its exchange. Following the failure of CGI across all its exchanges, Massachusetts decided to redesign its exchange instead of defaulting to the federal exchange.

After nearly 2 years and $1.3 billion of wasted money, the picture is clear for these 7 and all other remaining states – there needs to be a closer scrutiny of choices while choosing contractors for exchanges. With King vs. Burwell outcome, a lot of states might think of having their own exchange to keep their subsidies intact. When that happens, these costly mistakes and lessons will serve as a guidepost for other states. The answer will lie in a solution that does not repeat these mistakes and a provider that has a promising track record in running state exchanges.