Like a burlesque strip tease for health policy wonks, the slow motion unveiling of premiums for state health insurance exchanges has generated a lot of attention, unease, and, yes, excitement. The 2014 premiums, the first for Obamacare’s centerpiece feature of health insurance marketplaces, represent nothing short of a referendum on the “affordable” in the Affordable Care Act.
In just the past few weeks, Maryland and New York have joined the show. Putting aside the “rates are too high” vs. “rates are well below expectation” arguments, one fact seems obvious from looking at the rates. With the very big exception of California, the variance among plan rates is startling.
We took a look at the rates for silver plans, the ones most closely watched as they form the basis for computing the premium subsidies. We also looked at the variation between the second-highest and second-lowest cost silver-level plans. This is a simple way to eliminate outliers at the high end and the low end of the premium range. It also allows us to include the cost of the second-lowest cost silver plan, the premium from which the subsidies are determined.
Table 1 shows premiums that a nonsmoking 40-year-old would pay in four different major U.S. cities representing three different state-based marketplaces. In Baltimore, Maryland the difference in premiums between second-highest cost silver plan and the second-lowest cost silver plan is $119 per month, or 40 percent. In Maryland there are nine plans but four truly distinct issuers: Aetna, AllSavers, CareFirst, and Kaiser Permanente. (Coventry DE and Coventry L&H are subsidiaries of Coventry Health Care, which is a part of Aetna. CareFirst BlueChoice, CareFirst of Maryland, Inc. (CFMI), and Group Hospital and Medical Services Inc. (GHMSI) are part of the parent company, CareFirst, Inc.)
New York City premiums for the same 40-year-old vary by more that 75 percent, a total of $292 per month. (New York State requires full community rating, which means everyone buying the plan pays the same premium, regardless of age.) New York City has 16 unique issuers offering silver plans, including several Medicaid Managed Care Organizations and other new market entrants. In the New York market there are two general groups; traditional commercial carriers, such as Aetna, Oxford and United which all priced at approximately $600 or more a month. New entrants, like the Freelancers, New York Fidelis, and Metro Plus, all priced just over $300 a month or approximately half of the rate of the traditional carriers.
The contrast comes from California. In San Diego, the difference between the second-highest and second-lowest cost silver plans is just $20 per month, or 6 percent. In San Francisco, the difference is even less, just $10 per month, or 3 percent. There are five issuers in these two California markets, although there are eight products in San Diego and five products in San Francisco. California requiresall plans to offer a standard benefit package, which for silver plans including a $2,000 deductible, $65 for specialist visits and $25 for generic drugs.
Note for Table 1: See premiums for Maryland, New York, and California. The CMS standard age curvewas applied to derive a 40-year old premium from the published 50-year old premiums.
While it is not surprising that rates would vary among states and regions within a large state, it is surprising to see the large variance of rates within the same regional market at the same metal level. All plans in the same metal level within a state are required to offer the same ten essential benefits. They are also required to have the same actuarial value as computed using a federally created calculator. And, finally, they all will be playing by the same Minimum Loss Ratio rules that should limit not only administrative costs but also profits.
So, why, at the end of the day, do rates differ so greatly? And why is California different? We offer five reasons.
(1) Plan actuaries are working in the dark. Actuaries, a much maligned but needed profession, are integral to the process of plan rate setting. Their job is to look at the experience of a group and make an educated projection of what that group will cost to insure in the next future period. Actuaries have similar training and utilize similar tools to make predictions, and they are pretty good at it when they have solid information on the relevant group.
In the case of ACA exchanges, there are huge uncertainties about who is going to sign up and what their health history has been. Will the young and healthy join the exchanges in large numbers? How much pent-up demand for services will there be among those that have been uninsured? These are challenging questions with no clear answers in past experience. As a result, actuaries must make more assumptions than usual to estimate price and more assumptions lead to wider variation in results.
Understanding this issue, insurance executives may prefer a range of pricing estimates from their actuaries so that they can make strategic decisions about pricing within those broader parameters. It should not come as a surprise that some of these leaders have leaned conservative (rate high) while others have taken more risk (rate low). Indeed, the same executive might decide to price low in one area and high in another, based on where he or she most wants to grow business.
(2) Plan members are loyal. Based on years of experience with private insurance, Medicare Advantage, and Medicare Part D, insurers know that, once someone chooses a plan, they tend to stick with that plan. For example, according to CMS, during the first years of the drug program, just six percent of non-subsidized beneficiaries switched plans each year. Of course, this may represent satisfaction, but it also may represent market inertia, a widely recognized force in health insurance and many other markets. Inertia may be even more powerful in health insurance selection because a health insurer is an indirect provider of health services. People may be willing to retain an unsatisfactory health plan to retain access to their providers.
Regardless of the reasons, this stickiness of plan choice makes pricing below expected costs a potentially attractive first year business strategy. The long term objective of this strategy would be to accept early year losses in exchange for the enrollment of a large number of potentially loyal members. Because of inertia, plans could raise their rates for 2015 and beyond and count on many members reenrolling.
(3) Plan sponsors are not all the same. Some plan sponsors are already large players in their markets. They may already have attractive contracts with providers (hospitals, physicians, and other providers) based on controlling large amounts of business that they can build on to create exchange plans. In contrast, other smaller or newer players may not have the same leverage to gain similar deals.
Another variation on this theme is that Medicaid MCOs may be able to leverage their current Medicaid reimbursement rates, which typically are lower than commercial reimbursement rates, to gain a competitive advantage in the new exchange marketplace, particularly if their target market is the low-income population that will churn back and forth between Medicaid and exchange eligibility. Thus, one plan sponsor may have to charge higher premiums than another simply because their costs for providing the same benefits are higher.
(4) Plans are not same. While much may appear the same from one silver plan (or bronze, gold, or platinum) to another, there may be real differences that impact anticipated costs. These differences, in turn, can allow one plan to charge less than a competitor.
For example, while offering the same benefits, one plan may have a narrower provider network than another. A plan member’s choice of physician or hospital may be more limited. While the ACA has access requirements, the regulations provide flexibility for product offerings that rely on limited or select networks that can be more tightly managed for price and quality than large, open networks. The jury is still out on how receptive consumers will be to these products, but we can expect to see a good deal of experimentation with selective networks as a way to lower premiums and drive delivery system reform.
Another example is prescription drug benefits. While plans are required to cover roughly the same number of drugs as each other, there may be some big difference that impact costs. One plan may cover fewer brand name medicines than another, or one plan may place more drugs on higher cost sharing tiers than another.
(5) Active purchasing works. California’s rates tell a different story than rates elsewhere. In contrast to the 34 states where the federal government is operating the exchange as a clearinghouse that will accept all insurers who meet minimum standards, Covered California negotiated rates with each insurer with the implicit threat that the Exchange would exclude any insurer who did not come up with an acceptable rate. The success of this strategy in reducing rate variation may encourage more states to consider active purchasing in future years, though other states will continue to rely on market forces and a rate review system that requires actuarial justification for rate increases.
What does all this rate variation portend for the future of the ACA? The news is likely good. The marketplace will present potential plan members with clear choices within metal levels. Plans with relatively high premiums may choose to reduce premiums, or risk being eliminated by market forces. Alternatively, these plans may find a successful niche offering premium products to a segment of the population. Both outcomes would indicate success in market-based competition.
At a larger level, the availability of low-cost plans will help balance the exchange risk pools, potentially fueling a “health spiral” in which low rates attract young and healthy lives, which improves the risk pool and leads to better pricing in 2015, which further improves the risk pool and leads to even better rates in 2016 and beyond.
The experience of the Medicare drug benefit is useful. As with the ACA, the drug program began with a dearth of experience in knowing who would enroll. As with the rates we are seeing, actuaries and health plan executives had to make the same tough choices and the same calculations on capturing market share. As with the ACA, first year premiums varied greatly.
Over time, competition reduced, but did not eliminate, premium variation. Today, large variations still exist. However, not surprisingly, beneficiaries are largely voting with their wallets and flocking to plans that offer low premiums.
As with Part D, we can expect that plans with low premiums will attract the lion’s share of enrollment. Tax credits will accentuate this trend since they do not vary based on plan selection, meaning that consumers buying up or down will bear the full, marginal cost of any plan they select. If these plans can sustain these low premiums, they will maintain share over time. Competitors will be forced to reduce their premiums, exit the market, or pursue strategies that allow them to maintain profitability despite small market shares.
The other part of the story is that any state, like California, that chooses to play an active role in negotiating premiums with plan sponsors will speed these market forces along by eliminating outlier rates up front. Whether this strategy will, in the end, result in lower rates than allowing the market to operate on its own is something we’ll have to wait to see.
As the weeks go on, we’ll get more and more premium information from more states. In terms of rate variation, we can expect more of what we are already seeing.