America's Bitter Pill

by Steven Brill

  Filipic was working closely with David Simas, another campaign alumnus who had been assigned to supervise the White House’s Obamacare messaging and marketing, much the way he had worked on the communications effort during the congressional deliberations.

  But Filipic had another, more important day-to-day partner: a Chicago-based company called Civis Analytics.

  Founded by thirty-year-old Dan Wagner, the Obama 2012 campaign’s chief of data analytics, Civis was the home of the Obama campaign’s “big data” techies. During the 2012 campaign, Wagner and a team of coders and data analysts had developed tools that could sift data so finely that finding and keeping track of persuadable voters to make sure they turned out to vote was brought to a new level. Their data crunching had also enabled microtargeted advertising buys through social, online, and traditional radio and TV media.

  Soon after the campaign, Wagner and the group, which was now much celebrated in political and big data circles, had formed Civis to sell its services to nonprofits, governments, and companies. Its sole investor was Google chairman Eric Schmidt, who had helped organize their work as an informal Obama campaign adviser.

  The Civis website described its creation this way:

  Our company was born in a large backroom of the Obama 2012 reelection headquarters. We called it the analytics cave.… From millions of data points, we constructed the most accurate voter targeting models ever used in a national campaign. We predicted the election outcome in every battleground state within one point. And our work guided decision-making and resource optimization across the campaign.…

  This company is our next step: we are taking our team outside The Cave to solve the world’s biggest problems using Big Data.

  By the time of Sebelius’s testimony before Baucus’s committee, Enroll America was becoming one of Civis’s first and biggest clients, and the Obama White House’s best hope for avoiding Baucus’s train wreck.

  Civis would feed data to Filipic and her team, telling them where and how to reach the uninsured with their media messages, rallies, and door-to-door canvassing.

  Only this time not all votes were equal. Of particular concern were those who were called “the young invincibles”—those eighteen- to thirty-four-year-olds who didn’t worry as much about health insurance because they were relatively healthy and couldn’t imagine themselves being otherwise. They needed to be convinced to sign up because without them the insurance risk pools would be filled disproportionately with older, sicker people, which would cause premiums to go up, which, in turn, would cause still fewer of the young invincibles to sign up.

  The challenge of luring the invincibles was particularly worrying to Karen Ignagni and the insurance companies she represented because the age band that the final version of the law had stipulated was only three to one. This meant that insurers could charge the oldest applicants no more than three times what they could charge the youngest. That was better than the two to one ratio that had been in the original drafts of the bill, but not as good as the five to one the insurers had insisted was necessary to make the premium offering to the young invincible appealing.

  Yet the premium subsidies that would end up being available to most consumers would in many instances be so generous that the cost for a twenty-something to sign up would often be nearly negligible, depending on regional market factors. However, although the 2012 election was now behind them, the Obama team was still gun-shy about heralding those subsidies and the resulting low premiums, for fear of Obamacare being attacked as an income redistribution program. Instead, they pointed the press to how the reunion of the campaign’s data whiz kids was going to win the day.

  BULLETPROOF IN KENTUCKY

  In red state Kentucky, Governor Steven Beshear had no such concerns about touting the subsidies the federal government was offering. The Supreme Court’s decision the prior June had blocked Obamacare in one important respect: When it came to Medicaid expansion, the Court had ruled that, even though the federal government would pick up all of the cost for the first three years and gradually decrease its share only down to 90 percent in 2020, each state had to be able to choose whether to take the deal and expand Medicaid to all of its poor.

  In line with their party’s staunch opposition to anything related to Obamacare, for most Republican governors in red states, the choice was simple. They would reject the deal, despite the fact that the math seemed to dictate otherwise.

  For Beshear, a red state governor who was a Democrat, the calculus was different. Kentucky had 640,000 people—17.5 percent of its population—who were uninsured. They were living in a state with some of the nation’s worst health statistics. Of those 640,000 uninsured Kentuckians, 308,000 were living below poverty level and could receive the benefits of Obamacare only if Beshear expanded Medicaid.

  Beshear had already waded into the thicket of Medicaid in his state, because even the limited program that Kentucky then had in place was producing a crushing burden on the state budget. The year before, he had pushed through a cost-cutting reform of the limited program over the opposition of hospitals and other providers, who, because of Beshear’s changes, were now paid less per patient.

  So Beshear understood not only the need the program already filled and how much expansion would mean to his state, but also what the impact of Washington picking up the enormous cost of expanding it would be. Still, as he had told Carrie Banahan, he wanted to be sure he got the numbers right and had bulletproof evidence to counter what was certain to be Republican opposition.

  Soon after the Supreme Court’s decision giving him a choice, Beshear had commissioned a consultants’ study to do that math. The report from PricewaterhouseCoopers was unambiguous.

  On May 9, 2013, three weeks after the Baucus “train wreck” hearing, Beshear used the consultants’ white paper to justify his decision to expand Obamacare to reach every poor person in his state.

  “The expansion,” he declared at a press conference, “will help hundreds of thousands of Kentucky families, dramatically improve the state’s health, create nearly 17,000 new jobs and have a $15.6 billion positive economic impact on the state between its beginning in Fiscal Year 2014 and full implementation in Fiscal Year 2021.… In fact, if we don’t expand Medicaid, we will lose money.”

  “OSCAR” IS BORN

  On April 30, 2013, Joshua Kushner and his partners made a set of crucial decisions that were based on math that was not nearly as clear as the calculus presented by Governor Beshear. That was the day that the health insurance company they were starting from scratch on a cool-looking floor in the historic Puck Building (which his family owned) in lower Manhattan had to file papers with the state insurance department detailing the rates they were going to charge for their proposed bronze, silver, gold, and platinum plans.

  It was, Kushner later said, “pretty much a shot in the dark.”

  Not quite. They had spent hundreds of thousands of dollars on consultants who specialize in analyzing health data. They had done elaborate market research. And they had hired a company called MagnaCare that created networks of doctors, hospitals, and other providers for unions that ran their own insurance programs, and then negotiated the prices to be paid to those in the network.

  Kushner’s partner, Mario Schlosser—who had modeled trading strategies for a hedge fund and then worked at McKinsey—had crunched data he had bought related to fifteen million claims in the New York area spanning four years. Schlosser had programmed the data to weed out the doctors who billed too much for average results or engaged in excessive lab testing. So, they had kept all of MagnaCare’s network of hospitals—which was pretty much every hospital in the New York area—but not all of its doctors. And they planned to have their online concierge service encourage patients to use the most cost-effective half of the doctors that they did keep in the network.

  The data would constantly be updated and analyzed to do everything from evaluating physicians to predicting what kinds of preventive services or other int
erventions individual patients might need and then encouraging them to get it.

  By now, Kushner, Schlosser, and Kevin Nazemi (the third cofounder, who was overseeing the creation of the overall product and its branding and marketing) were beginning to have what looked like a real company. About six months before, in one of their brainstorming sessions they were struggling to figure out a name for the venture. They wanted something that would “humanize” health insurance. Kushner’s answer: “Oscar,” the name of his late great-grandfather.

  They had also hired a slew of suits—seasoned insurance executives who knew the landscape and were excited, rather than put off, by how Kushner and his partners disdained their industry as a bunch of knuckle draggers about to be shown up by the guys in the Puck Building.

  In January 2013, they had completed a license application with the state containing thousands of pages of details about their financial plans, their networks, and their suppliers.

  Then the guys from Oscar had begun to do what start-ups usually do first: seek outside investors.

  Kushner later told me that he had originally planned to have his own successful venture fund, Thrive, put in all the money. However, as he talked about Oscar with venture capital friends, some had asked about investing with him. Whether it was that or a desire to spread some of the risk, or both, by the beginning of 2013, Kushner and his partners had prepared a PowerPoint deck for potential investors that presented their vision for a health insurance company that was not really a health insurance company as the industry knew it.

  Oscar intended to outsource pretty much everything that conventional insurance companies did. MagnaCare would handle the network of doctors, hospitals, and clinics. A pharmacy benefit manager, Caremark, would handle prescription drug claims and the network of pharmacies to fulfill them.

  Another vendor, based in Ohio, would handle all claims processing, billing, and the issuing of Oscar’s stylized ID cards. It had agreed to charge $8.25 per member per month, plus a $150,000 setup fee. That meant that Oscar didn’t have to invest in the costly infrastructure that other insurers had, and that its costs would go up only as its number of customers went up.

  Separate visiting nurse and behavioral health (for illnesses such as drug abuse) services had been signed up, too.

  “Oscar’s assembly and orchestration of best-in-class vendors creates a uniquely asset-light and nimble insurer which is able to service members in novel ways,” the investor presentation declared.

  The data related to all of those outside services would be integrated so that the customer, and Oscar, had a complete, up-to-date dashboard of everything going on healthwise in the consumer’s life. Customers would get emails reminding them to fill prescriptions (and that if they used a generic they would pay nothing at all). Oscar’s medical staff would get an alert if a patient had called into the telemedicine service (which was also free) an abnormal number of times. Maybe something more serious was wrong, and the person should be encouraged to see a doctor rather than end up in the (more expensive) emergency room.

  Customers could log in and search for doctors in their network and see who was closest to their home or workplace. They could see who had what quality ratings (based mostly on a database of patient surveys Oscar had bought), and what the cost and their co-insurance payment would be. Other insurers had started trying to provide this kind of online information, but their efforts were clumsy, often incomplete, and hard to navigate. Oscar’s would be different.

  There was “huge leverage in data-driven optimization across the healthcare system,” the investor deck promised.

  And with Oscar having completed an expensive licensing process that presented “extremely high barriers to entry” for any other likeminded start-up, it was “now in a unique position to dominate a previously non-existent market.”

  That referred to the market in which individuals, rather than employers, bought health insurance. It was a market that had all but vanished in New York when the state passed a law that required that insurance companies not exclude people with preexisting conditions or charge them more, but that did not include a mandate that everyone had to buy insurance. That meant that people not provided insurance on the job would be inclined to buy it only if they had medical problems. As a result, premiums in the individual market had skyrocketed, precluding most New Yorkers from being able to buy insurance on their own.

  Now, with Obamacare, that mandate was in place. Better yet, Kushner’s investor deck noted, Oscar expected that 80 percent of its customers would be buying that insurance with Obamacare’s federal subsidies.

  Oscar would be competing for those customers, the investment presentation said, with “legacy” insurers who had “hard-to-read mailers, poor communication, no consumer-centric organizations, incompatible technology layers, and hundreds of confusing products.” Or, as Kushner told me, “We’ll be competing with companies who spend all of their time getting customers and then avoiding them once they get them.”

  With “one million uninsured” in New York City, it was a “$6B+ opportunity in New York City alone,” the opening page of the deck declared.

  However, the pages at the end—the ones that came after all the advantages of Oscar were celebrated—presented a less cheerful picture. Insurance, even Oscar-style, is not a Silicon Valley–or Silicon Alley–like home run (or strikeout) business. In the best of all worlds, with that medical loss ratio now in place, it still has to pay out 85 percent of the premiums it collects to healthcare suppliers, and then has to use the remaining 15 percent for all of its administrative costs before there is any profit.

  And unlike Kushner’s investment in start-ups such as Instagram, this one needed lots of money from the first day. Someone or something named Oscar can’t just collect premiums from people and promise to pay their medical bills if they get sick. State regulators don’t allow that. As condition of getting a license, an insurance company has to put sufficient funds on reserve to assure that those bills can be paid.

  When the state insurance department looked at Oscar’s initial plans it decided that it needed to deposit $30 million, based on a formula of one dollar in reserve for every four dollars of premiums Oscar projected it was going to get in 2014.

  Beyond that, the state required that all of Kushner’s budgeted $15 million in losses before Oscar was projected to break even had to be sitting in a bank.

  So that meant $45 million up front and in reserve, plus another $15 million for what would be spent in actual start-up costs for investments in people and infrastructure.

  Yet, a page at the end of the investor deck showed that in the group’s “conservative” projection, after depositing or spending that $60 million, Oscar would have just $100,000 in positive cash flow in 2016. The column showing the “achievable” projection said $10 million, and the one for the “imaginable” projection listed $45 million.

  A follow-on page showed much bigger numbers if Oscar could eventually roll out nationally. But most investors usually focused on the conservative projections—unless, like Kushner, they were as enthused about the idea of disrupting an industry they despised as they were about making a surefire bet.

  Ultimately, Kushner would get investments from several other venture firms, but his Thrive Capital would still be the dominant funder in the first year.

  Although the rates that Oscar filed on April 30, 2013, were based on reams of data analyzed by Schlosser and the actuaries Oscar had hired, and, therefore, hardly the “shot in the dark” that Kushner described, there was still a huge amount of guesswork involved, about which even the actuaries had little to go on.

  One example: Would the cost of Oscar’s unique, free telemedicine service, for which it would pay doctors to respond to callers within minutes on a per-call basis, actually result in no net cost because the telephone chats would head off so many expensive emergency room and doctors’ office visits? Their projection was that one in eight phone calls would avoid an “average health care visit”
with the cost of the visit averaged between high fees for an emergency room and lower costs for a doctor’s office. But maybe it would be one in four. Or one in ten. No one really knew. There was no precedent, no historical data to rely on. But unlike the Congressional Budget Office’s projections for Obamacare, Kushner and his team had to try to score it.

  A second example: Oscar’s advertising and general ethos was intended to target New York’s younger professional set. If it succeeded, Oscar would be capturing the young invincibles, whose healthcare costs would be lower than an average mix of young and older customers. However, Obamacare had a provision that required insurers who paid out less than was paid out by insurers with an average risk pool to pay into a fund to compensate insurers whose risk pool was above average. How was that going to work? The rules had not been set. Would that completely nullify Oscar’s advantage?

  Finally, there was the question of how anyone coming to the exchange would know how “cool” Oscar was. Could its logo be displayed? More important, it seemed as if the only information displayed on the New York State exchange’s opening comparison shopping page would be the premium and maybe the amount of the deductible. How would people know that Oscar customers got generic drugs, telemedicine, and other benefits for free?

  Yet Kushner, Schlosser, and Nazemi pressed on. Like all entrepreneurs they were sure they would think of ways to solve all that. Their basic idea, they thought, was just too good—too disruptive—not to succeed.

  CHAPTER 18

  TWO MONTHS TO GO

  July–August 2013

  IN JUNE 2013, I CAME UP WITH WHAT I THOUGHT WAS A GREAT IDEA for another special report for Time. As with that earlier story, this one was born of simple curiosity.

  I had gotten the idea for the first article when I became curious because so much of the debate over Obamacare seemed about how we could insure millions of Americans against the stupendously high cost of healthcare in the United States, rather than why the cost was so high to begin with. What and who was behind all of those high prices? Where was all that money going? To find out, I would trace the money and profits behind every line item in hundreds of pages of bills given to seven patients.