One of the ideas behind employee group policies and one of the purported virtues of employer sponsored health insurance is that the marketing and other overhead costs are lower. Instead of selling 200 policies to 200 persons, the insurer sells one policy to an employer. And one reason employer sponsored health insurance has often been supposed to be lower priced than comparable individual policies is that individuals who are gainfully employed tend not to have at least some of the expensive chronic or acute conditions such as certain forms of cancer or serious heart disease.
These assumptions predict that if one looks at policies sold in the individual exchange and the SHOP exchange under the Affordable Care Act that are sold in the same county, from the same issuer, have the same marketing name, have the same metal level, the same plan type (PPO, HMO, etc.), the same deductible and the same out-of-pocket limit, the SHOP premiums should on balance be lower than the individual premiums. Apples to apples, they should at least be no higher. Indeed, lower potential premiums are one of the key reasons for the existence of SHOP exchanges.
When I examine the data available from healthcare.gov, however, I find that the opposite is true.
SHOP policies are on average 8% more expensive than apparently identical individual policies.
That’s the key finding. The rest of this post tries to figure out why this might be the case. I’ll tell you the statistical ground I traversed in this effort. But I will confess that at the end of the day I am not quite sure why it is that SHOP premiums sure appear to be higher.
It’s not a matter of outliers skewing the mean. The median premium ratio between comparable SHOP and individual ratios — the “SHOP/individual ratio” — is 1.08. 10% of the comparable policies are more than 21% more expensive on the SHOP exchange. And even the lowest 10% are just 2% cheaper on the SHOP exchange than on the individual exchange.
The figure below shows the distribution of SHOP/individual ratios among the 12,689 comparable policies in the dataset. As can readily be seen, most policies are more expensive on the SHOP exchange. (A SHOP/individual ratio greater than 1 means that the median SHOP policy was more expensive than the median comparable individual policy).
Breaking down the data to gain insight
The state in which the policy was sold and the issuer might affect the SHOP/individual ratio. Unfortunately the effects of these potentially separate variables are extremely to tease apart because no issuer sold in more than one state. The table to the left shows the results. It shows significant variation in median SHOP/individual ratios across the combination of issuer and state. On the top end, Minuteman Health, which sells in New Hampshire has a ratio of 1.31 and Health Alliance Medical Plans, which sells in Illinois has a ratio of 1.25 Montana Health CO-OP has a ratio of 1.2. On the bottom end, CommunityCare, which sells in Oklahoma has a ratio of 0.76 and CoOpportunity Care, which sells in Iowa, has a ratio of 0.86.
The best I can do to tease out whether it is the state in which the policy is being sold or the issuer in the state that is responsible for the variation is to look at those states in which all the issuers have SHOP/individual ratios greater than the median of 1.08 and none have a lower one. Two states show up: Montana and Texas. This suggests some regulatory issue or special market condition in those two states that is leading SHOP premiums to be unusually high or individual premiums to be unusually low. Unfortunately, the finding is not particularly robust and I will confess to having little idea what this special factor might be.
We can see if the metal level and the plan type end up affecting the SHOP/individual ratio once the issuer and state are controlled for. Multivariate linear regression shows the impact of Metal Level and Plan Type to be quite small. The greatest effect was shown by POS plans, which reduced the SHOP/individual ratio by 0.03. Everything else had a smaller effect. Basically, SHOP plans tend to be more expensive than comparable individual market plans without regard to metal level or plan type.
An adverse selection theory?
Let me at least explore one other reason SHOP premiums might generally be higher. There are several buffers against adverse selection — the proclivity of persons with accurate knowledge of higher risk to select greater amounts of insurance coverage — in the individual market. A key one of these are the premium subsidies, which mean that many individuals, even those of relatively low risk, pay less for insurance than their expected cost of health care. There’s the individual mandate that likewise coaxes individuals, even those of low relative risk, to purchase insurance. There’s an open enrollment period. An individual can’t easily wait until they get sick and then by insurance in the middle of the year. In theory, they only can purchase insurance outside of the open enrollment window if they qualify for “special enrollment” by virtue of a small set of non-medical changes in their life circumstances.
The SHOP market may be more vulnerable to adverse selection problems. There’s no employer mandate that applies to small employers that tend to be eligible for SHOP policies. Although there are some tax credit subsidies, they tend to be less lavish than those bestowed on individuals and they apply only to some small employers. For other small employers there really aren’t any of the sort of “special deals” that might make it a good deal for even those whose employees and dependents are low risk. Finally there is no limited open enrollment period. On the contrary, under 45 C.F.R. § 155.725(b), “[t]he SHOP must permit a qualified employer to purchase coverage for its small group at any point during the year. ” Thus, an employer can wait until someone they care about — say the CEO’s daughter — gets really sick and then decide it would be prudent to have a employee group health plan.
What we may be seeing in the SHOP exchanges is an insurance world in which even the mild protections against adverse selection contained on the individual exchanges do not exist. And the result is predictable: higher prices and very few buyers.
So, what do we do with this finding? What do we do about the fact that SHOP policies tend to priced higher than comparable individual ones? Standing by itself, I am not sure one can draw much of an affirmative policy implication out of the result, particularly where we don’t yet have a good handle on causation. I do think it argues for thinking about imposing some adverse selection controls — such as limited open enrollment periods — if we are going to keep SHOP alive.
I also think, however, that the findings disclosed here have kind of a rebuttal value. They mean that proponents of SHOP exchanges should have a difficult time grounding an argument to preserve that additional complexity of Obamacare on grounds that it saves money. Although there are, to be sure, exceptions, and although there may be other reasons to induce small employers to purchase health insurance for their employees, right now it does not look as if the price is right.
Many have been concerned that the architecture of health insurance without medical underwriting created by the Affordable Care Act was inherently unstable and that, sooner or later, the markets it created would contract due to serious adverse selection problems. Although various creative bolsters from the Obama administration have delayed that forecast from yet materializing, except perhaps for the most generous of ACA exchange plans, as it turns out, the more immediate threat to Obamacare may come not from its inherent architectural deficiencies but from technical flaws now being unearthed by program detractors.
One of these flaws has been much in the news: the failure of the premium tax credits section of the ACA (section 36B of the Internal Revenue Code) to extend to policies sold in states that did not establish an exchange pursuant to section 1311 of the Act. There are approximately 34 such states. In 2014, they covered about two-thirds of those enrolled in individual health plans through the Exchanges. The Supreme Court is likely to decide this term in King v. Burwell whether the Obama administration’s determination to extend tax credits to persons in those 34 states is lawful. A decision against the Obama administration, which appears to be the prevailing prognostication, will throw major parts of the ACA into turmoil because only the sicker insureds with incomes that now qualify them for policies are likely to purchase those policies at full freight. Insurers, knowing of that proclivity, are going to be very leery of selling such policies; adverse selection would seem inevitable. It remains to be seen whether legislative action at the federal level — revision of section 1311 of the ACA — or at the state level — grudging creation of exchanges — would return those markets to equilibrium following a decision expected by many in King v. Burwell.
Another flaw, however, has not received much attention — until late. It is the apparent failure of Congress directly to appropriate money for another critical part of Obamacare that keeps premiums low: the cost sharing subsidies created by section 1412 of the law and now codified at 42 U.S.C. § 18071. The idea of this provision is that poorer purchasers can purchase a policy for “Silver” prices that ordinarily would have 30% cost sharing, but receive a policy that provides anywhere from “Silver plus” (27%) to “Platinum-plus” (6%) levels of cost sharing. This way, lower-middle-class people can get a policy that they might be able to afford without much of its purpose being undone by hefty deductibles and copays.
For the reasons I outline below, it appears clear that Congress at least strongly contemplated that provision of these extra benefits to the poor would come not from higher prices for policies paid by wealthier purchasers on the individual exchange. Instead, the federal treasury would pay the insurers for the extra costs they incurred in offering these more generous variants of the policy. It appears that the Obama administration has been making such payments to insurers, even if the amount of the payments — potentially in the billions — has not been made clear. (see 3:29:36 of this CSPAN video and the comments of CMS administrator Marilyn Tavenner).
In the lawsuit captioned United States House of Representatives v. Burwell, however, filed November 21, 2014, the plaintiff demonstrates with some care how Congress never actually appropriated any money for the cost sharing subsidies that sweeten Obamacare coverage. Presumably, insurers should thus have to cover themselves the resulting extra expenses created by higher utilization and lower deductibles and copay. Presumably insurers should do so out of revenues they receive from customers paying the full price. Gross premiums for everyone would thus need to be higher: probably 10-15% higher to cover the shortfall. And if insurers neglected to take those extra expenses into account, well, tough on the insurers one supposes. Such a lack of empathy would not be without recent precedent. Congress just hurt the insurers badly in section 227 of the Consolidated and Further Continuing Appropriations Act, 2015 (“Cromnibus”) by apparently cutting off a creative funding arrangement the Obama administration had undertaken to make payments to/bailout the insurance industry through the Risk Corridors subsidy program.
Incomplete funding of Risk Corridors is middling potatos, however, compared to non-funding of cost sharing. I would not be surprised to see an increase of 10-15% in gross premiums result if such cost sharing payments were found unlawful. An increase of (1) 10-15% resulting from the absence of appropriations for cost sharing subsidies, (2) perhaps 3% from whatever premium increases are likely to result from the “Cromnibus” decision not to permit circuitous funding of Risk Corridors deficits and (3) perhaps another 7% from increases in premiums that will result from the ACA-required phaseout of the Transitional Reinsurance provision under which the federal treasury covers insurers for insureds with large losses all adds up to a gloomy future for the Affordable Care Act. And that’s true even if, as its proponents claim, the cost curve is being bent. One reason insurance premiums are as low as they in the Exchanges is that, behind the scenes, the government is heavily subsidizing them in a variety of ways.
This cumulative projected increase can not be dismissed by asserting that the increase in premiums resulting from court-barred federal subsidies would affect only those earning more than 400% of the Federal Poverty Level and thus ineligible for Obamacare subsidies. Yes, it might appear that the net premium for others under section 36B really relates only to their incomes and not to the gross premium for insurance.
But the appearance of a limited effect is misleading in at least two respects. Increases in premiums resulting from court decisions and statutory reductions will matter more broadly. First, the subsidy only covers the cost of the second lowest silver plan in the rating area. The many people wanting a plan more expensive than that — a Silver PPO in many parts of the country or even a Gold or Platinum HMO — will be affected. Indeed, their net premiums will go up by a higher percentage than the increase in the gross premiums because the denominator of the increase calculation will not be the old gross premium but the (smaller) old net premium. Second , to the extent that insurers attempt to compensate for the premium revenue shortfall by raising premiums on employer-sponsored insurance, under Revenue Procedure 2014–37 (page 363), which purports to implement section 36B, such a move would trigger increases in percentages of income that individuals have to pay as the net premium for even the second lowest cost Silver Plan.
So, what’s the answer? We haven’t seen the literal answer in court to the complaint by the House of Representatives and, of course, there’s a very serious issue as to whether this is the kind of dispute that belongs in a court anyway. Bet the house that the Obama administration will raise issues called “standing” and “political question doctrine” in an effort to get the case dismissed. But, if those objections fail, is there an answer to the core of the House of Representative’s complaint on this point?
Congress intended that the federal treasury fund cost sharing
One answer might be that Congress at the time of the ACA’s passage clearly intended that payments for cost sharing reduction come out of the federal treasury and not through insurers charging higher prices. The evidence on this point seems rather compelling. Here is at least some of it.
In discussing premium tax credits and cost sharing reductions, Section 1412(a)(3) of the ACA says that the “The Secretary [of HHS], in consultation with the Secretary of the Treasury, shall establish a program under which— the Secretary of the Treasury makes advance payments of such credit or reductions to the issuers of the qualified health plans in order to reduce the premiums payable by individuals eligible for such credit. “
Section 1412(c), captioned “(c) PAYMENT OF PREMIUM TAX CREDITS AND COST-SHARING REDUCTIONS” states in subparagraph (3) “COST-SHARING REDUCTIONS.—The Secretary shall also notify the Secretary of the Treasury and the Exchange under paragraph (1) if an advance payment of the cost-sharing reductions under section 1402 is to be made to the issuer of any qualified health plan with respect to any individual enrolled
in the plan. The Secretary of the Treasury shall make such advance payment at such time and in such amount as the Secretary specifies in the notice. “
Section 1313(a)(6) of the ACA , captioned “APPLICATION OF THE FALSE CLAIMS ACT” states: “Payments made by, through, or in connection with an Exchange are subject to the False Claims Act (31 U.S.C. 3729 et seq.) if those payments include any Federal Funds. Compliance with the requirements of this Act concerning eligibility for a health insurance issuer to participate in the Exchange shall be a material condition of an issuer’s entitlement to receive payments, including payments of premium tax credits and
cost-sharing reductions, through the Exchange. ” This provision makes little sense if cost sharing reductions were not paid for by the federal government.
Section 1332 of the ACA addresses the possibility of states getting a waiver from many of the provisions of Title I of the ACA and says that in such event “the Secretary shall provide for an alternative means by which the aggregate amount of such credits or reductions that would have been paid on behalf of participants in the Exchanges established under this title had the State not received such waiver, shall be paid to the State for purposes of implementing the State plan under the waiver. ” Why would the State receive such funds for cost sharing reduction if the ACA did not contemplate that the federal government would already be paying for them?
Section 6055 of the ACA requires issuers of “minimum essential coverage” to provide information on the amount of any cost sharing reductions received. This provision makes no sense if insurers were just supposed to absorb the reductions and pass them on to other customers.
Section 10104(c) of the ACA addresses limits on use of federal funds to pay for abortions. It says no qualified health plan may pay for abortion services with “[a]ny cost-sharing reduction under section 1402 of the Patient Protection and Affordable Care Act (and the amount (if any) of the advance payment of the reduction under section 1412 of the Patient Protection and Affordable Care Act).” This prohibition would hardly seem necessary if cost sharing reductions were supposed to be absorbed internally by the insurer.
But perhaps it takes more than intent in a bill
My assumption, however, is that plaintiff House of Representatives will concede that the ACA certainly authorizes payments for cost sharing reductions and may indeed have contemplated that they would be made, but that it takes more than authorization for the executive branch. The House will argue, however, that to actually to make the payments: the Executive branch needs money. And it needs the money to be in the right account via a formal appropriation by Congress. The House will likely cite “The Purpose Statute,” 31. U.S.C. §1301 in support of this assertion. This statute reads: “Appropriations shall be applied only to the objects for which the appropriations were made except as otherwise provided by law. ” It will likely also cite 31 U.S.C. §1341(a)(1), the Antideficiency Act in support. It says “An officer or employee of the United States Government or of the District of Columbia government may not make or authorize an obligation exceeding an amount available in an appropriation or fund for the expenditure of obligation.”
These statutory citations are indeed foreshadowed by several paragraphs on the House complaint during which it recites the requests of HHS for appropriations to pay for Cost Sharing Reductions ($4 billion) and asserts that no such appropriation was ever made. The plaintiff notes that, by contrast, Congress did appropriate funds for the first cousin of Cost Sharing Reductions, advance premium tax credits through a standing appropriation under 31 U.S.C. § 1324 for tax refunds due individuals.
Or maybe not
I would expect two rejoinders to this argument. The first is a technical and statutory one: apparently the Secretary has at one time asserted that appropriations for premium tax credits also covers cost sharing reductions. The second is that any law restricting the executive’s power to spend money in this fashion is itself unconstitutional.
A statutory rejoinder?
Although acceptance of this first statutory argument would avoid the turmoil sure to erupt if cost sharing subsidies are judicially prohibited and the difficulties of constitutional adjudication, it strikes me, at least initially, as a loser. Although premium tax credits have a similar objective to cost sharing reductions, the two programs are not identical. They could operate independently. There are many who are entitled to premium tax credits who are not entitled to cost sharing reductions. If similarity of objective means that funds between programs are transferrable, an awful lot of Congress’ “Power of the Purse” has been evaded.
It’s also possible, however, since we haven’t seen the defendant’s response to the complaint that there’s some more authorization somewhere for the spending. If so, the House of Representatives is going to have egg on its face. I assume, however, that the House wouldn’t have been so foolish to file this lawsuit if it had not its homework carefully and failed to find even a needle-in-a-haystack explicit authorization for the spending.
A constitutional rejoinder
The harder question — and the one that would make House of Representatives v. Burwell a case about far more than the ACA — is the constitutional one. Under what circumstances does the President have authority to spend unappropriated funds? Much ink has been spilled by scholars on this issue over the decades . Tahere are some older Supreme Court cases (Hooe and Sutton and Bradley) that indirectly suggest that the limits created by the predecessors to these statutes are real and permissible. There’s also a thorough review of the then-existing literature by the Clinton-era Department of Justice in a memo of its Office of Legal Counsel from 2001 (2001 WL 36175929). Perhaps more relevant will be two cases which, though not binding on the Supreme Court, will likely have some precedential force.
Consider first Highland Falls-Fort Montgomery Central School District v. United States, 48 F.3d 1166 (Fed. Cir. 1995), a case decided by the Federal Circuit in 1995. It involved a statute, the “Impact Aid Act” designed to help certain categories of schools: (1) “section 237″ school districts whose property tax base was reduced by the presence of a lot of non-taxable federal property in the area, (2) school districts that had to educate children of workers on federal property, and (3) school districts that had incurred a substantial increase in the number of attending children. Highland Falls, which sits near the West Point Military Academy, was an example of the first kind of district. It should have received money pursuant to the Impact Aid Act since West Point apparently ate up apparently 50% — and a beautiful 50% at that — of the property in the district . But Congress, instead of allocating a lump sum for all payments to be made under the Impact Aid Act, split up the money with specific appropriation for each of the three categories of hardship it identified. And, apparently, the amount of money allocated to the category against which Highland Falls was claiming was short whereas the amount of money Congress had allocated to two other categories was more complete. So, Highland Falls wanted the Department of Education (DOE) to transfer money from the more fully funded accounts to the one that would benefit it.
The court in Highland Falls refused to direct such a reallocation of appropriated funds. Here’s what it said when DOE declined to do so:
Section 1341(a)(1)(A) makes it clear that an agency may not spend more money for a program than has been appropriated for that program, while § 1532 provides that an agency may use money appropriated for one program to fund another program only when authorized to do so by law. It is undisputed that, in each of the relevant fiscal years, Congress appropriated specific amounts to pay for § 237 entitlements. It also is undisputed that, in each of the relevant fiscal years, in order to fund § 237 entitlements at 100% levels, it would have been necessary for DOE to use money appropriated by Congress for entitlements under other sections of the Act—squarely in contravention of § 1532. The approach DOE followed was consistent with this statutory landscape.
As noted above, in order for DOE to fund § 237 entitlements at 100% in accordance with § 240(c), the agency would have had to transfer money from other sections’ appropriations to fund § 237. If DOE had followed such an approach, it would have been spending more money than Congress had appropriated for § 237 entitlements, in violation of § 1341(a)(1)(A). In addition, it would have been depriving at least one other section’s program of funds expressly appropriated for it by Congress. Put another way, it would have been “raiding” one appropriation account, for example § 238 or § 239, to credit another, § 237, in violation of § 1532.
Now, this is not a square holding on precisely the issue in the House of Representatives current lawsuit. It’s not a case where — as here — the Executive branch undertook a reallocation and someone wanted to challenge it. Nonetheless, the language of Highland Falls is supportive of the House’s point. Having decided, apparently, not to allocate funds for Cost Sharing, the executive branch can’t raid a related fund to help pay for it.
Also relevant will be Eastern Band of Cherokee Indians v. United States, 16 Cl. Ct. 75 (1988). There an Indian tribe sought money to equalize funding of its schools relative to local schools. There was a federal statute that was supposed to provide such money. But Congress had declined to appropriate funds for this special “set aside.” The tribe asked that money be used from other accounts controlled by the Secretary of the Interior to make the statutory payments. The court upheld the government’s decision not to do so.
The Set–Aside Fund was not funded in fiscal year 1986, the year of plaintiffs’ request. Plaintiffs argue that the Department of Interior could have applied funds from other accounts. However, the Anti–Deficiency Act, 31 U.S.C. § 1341(a) states that a United States officer may not authorize expenditures “exceeding the amount available in an appropriation or fund for expenditure or obligation.” Thus, the officers of the Department of the Interior could not grant the plaintiffs’ request for funding. Penalties for violating the Anti–Deficiency Act are codified at 31 U.S.C. §§ 1349 and 1350. The court thus finds that the plaintiffs have failed to state a claim upon which relief may be granted as funds are not available to satisfy plaintiffs’ claim.
Again, not a case 100% on point, but still one that, at least in dicta, reinforced the House’s claim here that the executive can not dip into one pot of money, even if related and even if efficient, to pay bills for another program. And that is true even if Congress has earlier expressed its intent that such a program be funded.
And there is, on the other hand a case involving a disappointed bidder and military purchases of diced turkey (with gravy) and beef stew: Southern Packaging and Storage Company, Inc. v. United States (D.S.C. 1984). There, a district court found that, although the purchase from a Canadian company violated the “Buy-American” provision of the Department of Defense Appropriations Act there was no violation of the anti-deficiency statute because the amount spent on combat rations — even Canadian-sourced ones — did not exceed the overall Congressional appropriation.
There is, in addition, lots of non-judicial authority on the subject, ranging from death-match law review articles by Professors Sidak (1989 Duke L.J. 1162 (1989)) and Stith, (97 Yale L.J. (1988)), to summaries of the law from the United States General Accountability Office to a memorandum from the Clinton-era Justice Department.
So, there is a lot more to be said on this subject and we have not yet had the benefit of Secretary Burwell’s research and argument. But, at least for now, provided the House can overcome the substantial justiciability questions, it looks like it may have a strong case on the merits. Of course, the House ought, like all of us, to be careful what it wishes for. One wonders what reaction many Americans will have to a House legal victory when they find that they can no longer afford the health insurance they purchased due to what they may well regard as a “technicality.”
Let’s start with some facts we can all presumably agree on. MIT Professor Jonathan Gruber was involved in the development of the Affordable Care Act. He attended numerous meetings with the executive branch officials while the ACA was being formulated, met with President Obama once, and stayed as a member of a Congressional Budget Office Advisory Council on Long Term Modeling for a decade, including the years when the ACA was designed. Although perhaps he exaggerated in an effort to draw attention to himself, he referred to himself, as others did, as the architect of Obamacare. Although he is hardly President Obama himself or Senators Harry Reid, Max Baucus or then Speaker of the House of Representatives Nancy Pelosi, Professor Gruber was not a mere technocrat crunching numbers. He is more intimately connected with the bill, more of an insider, than many other academic proponents of the legislation. And so he has described himself.
So, when we are looking to understand a challenging provision of the ACA, and if we accept that the provision is sufficiently ambiguous (in context) to be subject to broader interpretive methods, and if there isn’t much other contemporaneous evidence on the subject, it does not become crazy to look at Professor Gruber’s statements about the provision.
The provision of which I speak is, of course, is section 36B of the Internal Revenue Code (section 1421 of the ACA) in which, to the untrained eye, Congress appeared to limit advance premium tax credits (subsidies) to those “enrolled in through an Exchange established by the State under section 1311.” Only 16 or so states established such an Exchange. The remaining 34 in one form or another have let the work be done by the Federally Facilitated Marketplace established in section 1321 of the ACA as a fallback precisely when the States did not, as anticipated, establish an exchange. But the IRS has interpreted “established by the State under section 1311″ to include exchanges “established” by State non-establishment, i.e. their not establishing an Exchange knowing that the federal government would do so for them. This latter interpretation means that, just because people live in states with entrenched opposition to the ACA, like Texas, or states which have recognized their apparent incompetence in running an Exchange, like Oregon, or other states, which perhaps didn’t want the trouble, they will not be denied thousands of dollars of subsidies in a program which, at least according to the rhetoric of its proponents, was intended to reduce the rank of uninsured nationwide.
This provision, section 36B, is one that presumably the Supreme Court will interpret this term in King v. Burwell — unless of course it “DIGs” the case and decides to withdraw review for now. Although Burwell is not a constitutional case, and although it may have few jurisprudential ramifications, from a practical perspective, it is an extremely important decision. Because of the way section 36B reads, it is likely to determine whether many millions of Americans who have purchased health insurance on the “Federally Facilitated Marketplace” (FFM) pursuant to Obamacare in reliance on an advance of federal tax credits are in fact entitled to those advances or tax credits at all. It is about whether insurers will continue to sell health insurance policies in states now served by the FFM or seek to withdraw for fear of unsubsidized policies being bought predominantly by those with high projected medical expenses. And it is about whether some states will be induced by a decision in King v. Burwell to mitigate the damage to many of its citizens that would otherwise occur, by now establishing Exchanges whose creation they previously opposed. Oh, and if subsidies end up being unavailable, the employer mandate (26 USC 4980H) could be diluted because few employees will actually purchase policies on an Exchange.
It’s also about politics. The Democrats may look bad for having misused executive power to stretch the interpretation of an arguably clear law beyond recognition. And, of course the collateral political consequences of a “win” by mostly Republican opponents of Obamacare at the Supreme Court may provide that party with the credibility that comes from having a litigation position vindicated by the nation’s highest court. But all will not end there. At least some of this perceived political advantage to the GOP may be offset by the political harm likely to occur if the “victory” rips health insurance from their constituents. And it augurs a delightful spectacle: Republicans joining Democrats in the aftermath of the former’s victory in King v. Burwell to amend the ACA to actually say what the Obama administration now says it means. One can hear now Republicans claiming that it was all a matter of principles, of defending separation of powers and the Rule of Law. One could also perhaps see some Republicans wanting to take such a victory as a hostage and seeking concession from Democrats on a variety of matters, including a renegotiation of many provisions of Obamacare, as a condition of restoring coverage to millions of Americans.
Did Gruber lie?
But back to Gruber. The problem for those who support the IRS’ interpretation of section 36B is that it takes a heroic stretch of statutory language to get there. And Gruber — on videotape — twice — offered what purported to be a knowledgeable account of at least a plausible reason why the drafters of the ACA might have indeed threatened to punish the uninsured in states unwilling to “get with the program” and establish Exchanges. You can watch him below starting at about 31:25. Here’s a transcript.
Question: You mentioned the health information exchanges through the states and it’s my understanding that if states don’t provide them the federal government will provide them.
Gruber : Yes so these health insurance exchanges … will be these new shopping places and they’ll be the place that people go to get their subsidies for health insurance. In the law it says that if the states don’t provide them the federal backstop will. The federal government has been kind of slow in putting in the backstop I think partly because they want to sort of squeeze the states to do it. I think what’s important to remember politically about this is that if you’re a state and you don’t set up an exchange that means your citizens don’t get their tax credits. But your citizens still pay the taxes for this bill. So you’re essentially saying to your citizens, you’re going to pay all this taxes to help all the other states in the country. I hope that that’s a blatant enough political reality that states will get their act together and realize there are billions of dollars at stake here in setting up these exchanges and that they’ll do it. But, you know, once again the politics can get ugly around this.
Or here. It’s an audio from January 10, 2012 at the Jewish Community Center of San Francisco. Again, here’s a transcript
I guess I’m enough of a believer in democracy to think that when the voters in states see that by not setting up an exchange the politicians of the state are costing state residents hundreds of millions and billions of dollars, that they’ll eventually throw the guys out. But I don’t know that for sure. And that is really the ultimate threat and that is will people understand that, gee, if your government doesn’t set up an exchange you’re losing hundreds of millions of dollars in tax credits to be delivered to your citizens. So that’s the other threat: will states do what they need to do to set it up.
Per Gruber, it was all a bluff. It was a stick to get the states to establish their own Exchanges. It’s not all that much different from lots of conditional spending decisions, such as tying federal highway funds to state raising of the drinking age, except this was a conditional taxing decision. It’s a not-so-unusual way of nicely inducing the states to do something they might otherwise be reluctant to do because, now, not doing so, hurts their citizens. As careful health law scholar and Obamacare advocate Tim Jost pointed out in a 2009 article (see figure below), conditioning subsidies on the state’s creation of an exchange is a way around a potential constitutional impediment to simply directing that the states do so. And if the states call the bluff, well, so be it, that’s a matter for internal state politics and does not undercut the federal desire that the states behave in conformity with the incentives. The limitation on subsidies set forth by the text of section 36B was a stick so big and so bad that resistance was thought to have been futile. Indeed, that may well have been why Professor Gruber, as he stated under oath in his testimony this week before the House Oversight Committee, always assumed in his modeling that subsidies would be available in all states.
There was only one problem. Many of the states called the statute’s bluff. They refused to establish their own Exchanges, either seeking to avoid the financial obligation or, at least in the Red Zone, complicity with the evils of the Affordable Care Act. And so, having seen the bluff called, the IRS, under this theory, pretended that the statute had never conditioned subsidies on state creation of an Exchange. In order that the benefits of Obamacare extend from sea to shining sea, the IRS interpreted “established by a state under section 1311″ to include inaction by a state under section 1311 that led, under section 1321, for the federal government to come to the rescue.
Now, in ordinary circumstances, the musings, even recorded musings, of a lone professor at an academic conference or a community group on why Congress might have written a statute which, if one believes in many of the ideas of the ACA, is rather cruel, would not be particularly relevant to a Supreme Court case on its interpretation. After all, even careful law review articles by scholars are frequently ignored in statutory debate. And there is even a respectable argument that Gruber’s remarks are not relevant now to King v Burwell.
But interpretation disdains a vacuum. And the problem is that none of the legislators apparently explicitly focused on the purported cruelty of a literal interpretation of section 36B at the time the ACA was pushed through. And their silence could be interpreted several ways: that most people who cared understood it was a bluff that likely would not be called, that most people who cared understood that “established by a state under section 1311″ should be read broadly, or, perhaps most realistically, that most had no idea about the details of a 2,700 page bill, even one that had indeed been widely debated.
And so, if the executive branch is to prevail in its reading of section 36B, it would sure help if it could tamp down contemporaneous evidence some proponents, even non-legislative ones, thought that use of a bluff made any sense. Professor Gruber’s comments, as an important proponent of the ACA, thus acquire additional saliency.
To be sure, Professor Gruber at the same hearing before the House Oversight Committee had an explanation for his assertions on this point. It was one, I assume he and others hoped, that would further diminish the force of what he had to say earlier on. Its an argument based on allegedly omitted context. Here is what he had to say (go to about minute 34 of the CSPAN video):
About my January 2012 remarks concerning the availability of tax credits in states that did not set up their own health insurance exchanges: the portion of these remarks that has received so much attention lately omits a critical component of the context in which I was speaking. The point I believe I was making was about the possibility that the federal government for whatever reason might not create a federal exchange. If that were to occur and only in that context then the only way that states could guarantee that their citizens would receive tax credits would be to set up their own exchange.
In other words, Gruber now claims that the only circumstance under which citizens of states not setting up their own exchanges would be deprived of tax credits would be if the federal government did not set up an exchange either. In that event, even under the broad definition of “established by a state under section 1311″ that he embraces, there would be no exchange and the citizens would lose out.
The main problem with Gruber’s remarks is that the purportedly clarifying context is invisible except retrospectively and in Gruber’s own mind. Nowhere in his answers — nowhere in the full recordings — does he indicate a belief that Washington would not set up an exchange at all – a reasonable omission given that Washington was very much in the throes of establishing a federal exchange at the time. Washington spent hundreds of millions on healthcare.gov but was never going to get it up and running?
Want more evidence of the absurdity of the hypothetical scenario created by Gruber to reconcile his earlier comments with the desires of the Obama administration in King v. Burwell? You could read this May, 2012 report from CMS in which it discusses over 19 pages how the federally facilitated marketplace will work. You could read these July 2011 regulations and find the eight places in which the Department of Health and Human Services set forth how it is going to set up a federally facilitated marketplace, including the passage in the figure. Find me the warnings from CMS, from HHS, from the President from anyone that, in fact, the federal government was not going to establish a federally facilitated marketplace.
Not good enough? How about contemporaneous words very close to Gruber himself. Take a look at the work in December 2011 of the Study Panel on Health Insurance Exchanges. It’s important not only because it was work mandated by Congress but because a member of that study panel was … Jonathan Gruber. (Look at the list of panel members on page ii.) It writes a 34-page report on precisely how the federally facilitated exchange — the thing Gruber now says he doubted might exist — would come into being and the steps already being taken. The figure below is an excerpt from page 12 of that report.
It is thus no surprise that the report ends with the following statement: “Over the next 12 months, the federal government will continue to invest in and build a Federally-facilitated Exchange to operate in states that elect not to operate a State Exchange, or are unable to meet the certification and implementation schedule to stand up their Exchanges in 2014. ” In short the hypothetical scenario set forth by Professor Gruber in which the federal exchange does not exist looks like a fantastic reconstruction of events that simply did not occur.
And what are we to make of Gruber’s “squeezing the states” language? Are we to believe that the federal government thought tax credits were so important for a nationwide program that they would squeeze the states by going slowly on establishing an exchange only then to not set up an exchange at all if some states failed to capitulate to the pressure? How would that be consistent with a belief that the ACA was supposed to establish a nationwide program? And what are we to make of his language about state democracy: “I’m enough of a believer in democracy to think that when voters in states see that by not setting up an exchange the politicians of the state are costing state residents hundreds of millions and billions of dollars, that they’ll eventually throw the guys out.” It wasn’t the federal officials, his Obama administration friends, that Gruber hoped the state voters would throw out for failure to establish a backstop exchange; it was the officials in the state that he hoped would be thrown out for failing to establish an exchange. The simplest explanation for this hope is that Gruber believed that under the statute, even if the federal government established an exchange that let people buy policies without subsidies, states not establishing their own exchanges would thereby cause their citizens to lose hundreds of millions of dollars in tax credits.
I’m afraid he did
In short, and I say this with some sorrow as a fellow professor who has testified before the same committee, there is proof beyond most doubt that Professor Gruber deliberately lied under oath on at least this point. He did so not in an off the cuff remark but with advice of counsel and after having apparently rehearsed and written out his testimony beforehand.
Now, to be complete, I suppose we order to consider one other make-weight explanation offered by Professor Gruber to diminish the import of his earlier recorded comments: it’s about his models.
Indeed, my microsimulation models for the ACA expressly modeled that the citizens of all states would be eligible for tax credits whether served directly by a state exchange or by federal exchange.
But this explanation about his behavior presumably prior to the enactment of the ACA is not inconsistent with a view that 36B conditions subsidies on a state creating an exchange. It’s consistent with a (mistaken, as it turned out) view that the carrot and stick contained in section 36B was too large for states to ignore. It is also potentially inconsistent with his belief, expressed two sentences earlier, that he, unlike anyone else in the debate, harbored this suspicion that Washington would not set up an exchange for the states that failed to set up their own. In that event, according to Gruber’s own reasoning, he should not have assumed in his model that all states would receive subsidies.
Why does it matter?
Ok, so some MIT professor interpreted section 36B of the ACA the way the plaintiffs in King v. Burwell do. OK, so he took liberties with the truth in his testimony before Congress. Is this an irrelevant tempest in a teapot brewed up by implacable Republican adversaries of Obamacare? I don’t want to speculate on motivation, but I actually think it is neither the most important event in the history of Obamacare — far from it — nor entirely irrelevant. It may well be that the statute is so clear, though, that Professor Gruber (or anyone else’s thoughts) on its meaning are entirely beside the point.
Nonetheless, if for no other reason than to satisfy curiosity, I would like to see Professor Gruber, when he is hauled back before Congress pursuant to an additional subpoena, asked a more open ended question about how he, a mere (MIT) economics professor without legal training acquired his beliefs about the meaning of section 36B. Did he really read the statute with care and come to that conclusion using the same somewhat — let us be fair — circuitous statutory reasoning now advanced by the defendants in King v. Burwell? Or might it have actually been based on comments he heard from the true legislative architects of the ACA during some of the many meetings he held on the subject? If so, even such hearsay might be more relevant than Gruber’s own beliefs as to interpretation of a critical statute.
I am also old fashioned enough to be somewhat concerned about what sure looks to me like at least one calculated lie. If, for example, Professor Gruber believes so strongly in the ACA — and one need only read his graphic book to realize the passion of his commitment — that he is willing to re-invent events in order to play a tiny role in its salvation, how non-instrumental was he in the modeling that led up to passage of the ACA and that, I believe, may have some residual role in contemporary forecasts of its success? I can understand that lies in order to provide, in his opinion, millions of people access to life-saving healthcare may in his mind be a necessary evil. After all, although transparency may be important, it is crystal clear that Gruber would, as he said, rather have this law than not.
A key feature of the Affordable Care Act are the “SHOP Exchanges.” These are markets established by either a state exchange pursuant to section 1311 of the ACA or the federally facilitated market (FFM) pursuant to section 1321 of the ACA in which employers with fewer than 50 full time equivalent employees can purchase health insurance for their workers. They can do so without being subject to either rating based on the projected health of those covered or “experience rating” in which premiums are tied to health expenses in prior years. SHOP exchanges are intended to be a mechanism whereby the burden of providing healthcare can remain where some people (not me) believe it belongs: with the employer.
To date, the key feature of the SHOP exchanges has been their complete failure to attract customers. A GAO report issued in November, 2014, said that only 76,000 individuals—including employees, their spouses, and dependent children— had enrolled in SHOPs operated by a state and, although the data was, amazingly enough, not available from the federal government for the remaining states whose marketplace it operated, enrollment had apparently followed a similar dismal pattern. This is in stark contrast to yet another stunningly wrong forecast of the Congressional Budget Office, which, as late as April 2014, had made forecasts that tax credits under section 45R of the Tax Code would cost $1 billion for 2014, an estimate that implies enrollment by more than a million, persons in the SHOP exchanges.
There have, of course, been multiple explanations of the failure of the SHOP exchanges in 2014. As has been well documented, the computer systems for the Federally Facilitated Marketplace (FFM) simply did not work. Applicants thus had to resort to more primitive paper processes. Second, the transient nature of a tax credit offered to some SHOP purchasers may have been insufficient to attract buyers, some of whom likely feared it would heighten expectations among employees of employer provided coverage that would be difficult to sustain in two years when the tax credit expired. And a variety of other explanations have been offered ranging from insufficient advertising to competition from non-SHOP-exchange markets, to difficulties with use of traditional intermediaries in the new market.
But perhaps, as some have noted, the basic problem is that it purchase of policies on a SHOP exchange at full or close-to-full price just doesn’t make doesn’t make a lot of economic sense when it appears that most employees could otherwise qualify for a taxpayer funded subsidy if they purchased similar policies on the individual exchanges. If small employers have the funds to help their employees more and want to help them meet medical expenses, perhaps the best medicine would be green: cash. Such a choice would have the fringe consequence — perhaps fringe benefit — of pushing more healthy people into the individual exchanges thereby reducing the risk of an adverse selection death spiral. It might also have distributional consequences that many would regard as an improvement. One can thus see the seed of bipartisan support for repeal of this provision.
The rest of this entry explores this proposition using four sample scenarios. Given that employees can purchase policies on the individual exchange that, at least for now, will be subsidized by the federal government in many instances, does it make sense for benevolent employers to provide health insurance or to instead, give workers higher wages, and let them choose to purchase policies that may be subsidized on the individual exchange?
Several factors matter in determining whether it makes sense to SHOP from the perspective of an employer and employee:
1. How much of a subsidy, if any, is the employee entitled to if it purchases a policy on the individual exchange. This may in turn depend not just on the compensation the employer pays the employee but also on the size of the employee’s household and sources of income of other household members. The higher the subsidy, the less generally it is in the interest of the employee that the employer purchase coverage through the SHOP exchange.
2. Would the small employer be eligible for a tax subsidy under section 45R of the Tax Code (section 1421 of the ACA).?These subsidies can range up to 50% of premiums. Generally, the higher the 45R subsidization rate, the greater the possibility that it is in an employee’s interest that the employer purchase coverage through a SHOP exchange rather than provide the employee with cash.
3. How do premiums in the individual market match up with premiums in the SHOP market for comparable policies. Many supposed that SHOP market policies might be cheaper and provide an advantage to employer purchase. But to the extent there are no such savings or, as perhaps is turning out to be the case, SHOP policies are more expensive, the case for having a SHOP exchange option is weakened.
4. To what extent do the preferences of the employer about which plan to select (Metal Levels and Plan Types) match the preferences of the employees. To the extent the fit is poor, that is factor suggesting that benevolent employers do better to avoid the SHOP exchange and instead give their employees cash so they may do as they see fit.
If it turns out that few employees benefit from the existence of the SHOP exchanges or that its distributional consequences are not sensible, the case for simplifying the Affordable Care Act by elimination of this component makes some sense.
Here are the families in the scenarios: Dora, the Dursley family, James and Mary, and Ada.
Example 1: Dora
Consider Dora, a 40 year old single woman making $27,195 per year working for a business Exploration, Inc. , that employs 40 full time equivalent workers and is thus ineligible for any sort of section 45R subsidy. For concreteness, we’ll put Dora in Harris County, Texas (Houston). If Exploration, Inc. goes to the SHOP marketplace for 2015, it will find that the second cheapest silver plan is sold by Blue Cross for $400 per month ($4,800 per year) for a single person age 40. The policy features a $3,000 all-inclusive deductible and a $6,350 all-inclusive out-of-pocket limit. So, if Exploration, Inc. were to purchase this policy, Dora would have health insurance and no additional tax liability as a result.
Suppose that in lieu of paying $4,800 in premiums to buy health insurance for Dora, Exploration just raises Dora’s wage by $4,800 . Call this her gross “in lieu compensation.” Assuming Dora is in the 15% marginal tax bracket, she would actually receive $4,080 of in-lieu compensation from her employer. If Dora then went to the federally facilitated marketplace for Texas, Dora would find that she could purchase an essentially identical policy from Blue Cross with the same deductible and out-of-pocket limit for $363.60 per month or $4,363 per year as a gross premium. But, at least until a decision against the Obama administration in King v. Burwell throws the nation into chaos , Dora would be eligible for a small subsidy from the federal government of $15.22 per month since the second cheapest silver plan in her area (a “Blue Advantage Silver HMO”) costs $250 per year. Dora’s net premium would thus be about $4,181.
Thus, to be in almost exactly the same position as would have been had Exploration purchased a policy on the SHOP exchange, Dora would be out a net of $101. Maybe, at least for Dora, its marginally in her interest if her employer goes SHOPping.
But this small negative sum may well be offset by a benefit Dora receives as a result of Exploration not going to the SHOP exchange and not buying health insurance for its employees: freedom. Freedom could help Dora in a variety of circumstances.
Suppose, for example, that Dora is in great health, has some money saved up, and just wants a Bronze policy to cover her against catastrophic medical expenses. Now Dora can go to the Exchange and pick up a Blue Cross Bronze “Blue Advantage Bronze HMO” policy for just $191.03 per month gross and about $176.03 per month after her subsidy. After she pays this net premium and her taxes on the $4,800 she got from her employer, she’ll have about $1,970 left in her pocket. If her medical expenses for the year end up being $4,970 or less for the year she’s come out ahead relative to where she would have been under the hypothesized Silver policy purchased by Exploration.
Or, suppose Dora isn’t in such good health and doesn’t like the risk created by a silver plan. She wants a platinum plan. She could obtain the “UnitedHealthcare Platinum Compass 250″ plan with a deductible of just $250 and a out-of-pocket limit of $1,500 for a gross premium of $328.91 per month or $3,947 per year. The net premium for the policy after consideration of subsidies would be about $3,764 per year. She could buy this with the extra compensation she received from Exploration Inc. and still have $316 left over. If her medical expenses ended up being very low or somewhere between $1500 and $6,350 she will end up ahead.
Example 2: The Dursley Family
Vernon Dursley is the 40-year old Vice President Grunnings, a small drill manufacturer in Potter County, South Dakota. His family consists of his 40-year old wife Petunia and his young son Dudley, who is diabetic. Grunnings employs 15 individuals full time who have an average annual wage of $23,000. Dursley himself earns $90,000 per year. If Grunnings went to the SHOP exchange operating by the FFM for South Dakota, it would find a Bronze plan, “Sanford Simplicity-$3,000″ available at a gross premium of $716.05 per month to cover the Dursley family. Because, however, Grunnings employs fewer than 25 people and pays a low annual average wage, it is eligible under section 45R of the Tax Code (section 1421 of the ACA) for a federal tax credit for one third of the amount of such purchases. Thus, the amount Grunnings would save by not purchasing the policy — at least with respect to Mr. Dursley — is $477.37 per month or about $5728 per year.
Giving Mr. Dursley $5,728, will not, however, be enough to enable him to purchase comparable coverage for his family on the FFM for individual policies for South Dakota. He is not eligible for a subsidy because his income is more than 400% of the applicable federal poverty level. First, the Dursley’s marginal tax rate is likely to be 25%. Thus, he will not net $5,728 from the Grunnings bonus; he will net $4,296. The least expensive Bronze policy for the Dursleys would be the “Dakota Reserve 6000″ for $558.58 per month or about $6,703 per year. Thus, the Dursleys would be about $2,407 worse off if Grunnings failed to purchase a SHOP policy.
The underlying reason that the Dursleys do better if Grunnings SHOPs is that that Mr. Dursley is a well compensated employee of a company that is eligible for a tax credit if it goes to the SHOP exchange. Thus, even if premiums in the individual market are a little lower for comparable benefit packages as they are in the SHOP market, that benefit ends up being overwhelmed by the differential tax treatment.
Example 3: James and Mary
This example is somewhat more complicated but it is also quite important.
James is a 60 year old worker in medium health working for Peaches Unlimited in Peach County, Georgia. He’s married to his 60 year old wife, Mary. His household makes $50,000 per year. Peaches, which has 30 full time equivalent employees, has considered going to the SHOP Exchange and providing Bronze coverage for its employees and spouses, paying for all of the employee’s premium expenses and for half of the spouses. When it does so, it finds one plan: “BCBSHP Bronze Pathway X Enhanced 5000 30 6600 Plus” a POS (point of service) plan from the Georgia Blue Cross/Blue Shield carrier. The premium attributable to a 60 year old couple is $14,652. But Peaches will only pay half the premium for spouses, so, if Peaches were to purchase this policy, the annual incremental cost attributable to James and Mary would be $10,989. For this, James and Mary would get a plan with a $10,000 deductible for medical expenses, an additional $1,000 deductible for drugs, and a $13,200 out-of-pocket limit. If they wanted the policy, however, James and Mary would have to cover half of Mary’s premium, which would add an additional $3,663 to their costs. So, if we are to compare apples to apples, we need to see what James and Mary’s financial position would be if they shopped instead on the individual exchange.
Suppose that Peaches gives James and Mary $10,989 in in lieu compensation. They would likely have a marginal tax rate of 15%, which would make their after-tax additional compensation about $9,341. If James and Mary try to acquire a matching plan, the closest it looks they can come is the Humana Bronze 6300/National POS – OpenAccess plan, which has a gross premium of $1,181 per month or a hefty $14,177 per year. The net premium, however, will be only $5,578 after premium tax credits are received. Thus, James and Mary will be able to take the $9,341 in after-tax compensation they received, pay $5,578 and have $3,762 left over. But it’s better than that. James and Mary won’t have to pay $3,663 for half of Mary’s policy. So, in fact they will be $7,426 better off and Peaches no worse off if Peaches does not go to the SHOP exchange. For a couple making $50,000 that is a lot of money.
How has this happened? It is the result of three factors coming together: (1) James and Mary being eligible for a large subsidy from the federal government due to their age and the consequent high cost of a silver policy; (2) Peaches having too many employees to be eligible for any sort of tax benefit from the government; and (3) the high price of SHOP policies in Peach County relative to individual policies. As shown here, when these factors come together, employees should prefer a situation in which employers give them cash, not expensive health insurance policies. Moreover, if Peaches does not go to the SHOP exchange, James and Mary are no longer wedded to the likely small number of plans Peaches wants; they can pick any plan that best meets their needs available in the individual exchange.
Ready for one more?
Example 4: Ada
Ada is a 30 year old programmer at Babbage Enterprises , a small tech firm in Allegheny County, Pennsylvania. It has 10 employees but their average wage is $75,000 per year, making Babbage ineligible for any section 45R tax credit. Ada is single and makes $100,000 a year. It believes employees should have the best health care available. Babbage, which has a CEO with recurrent heart problems, could go to the SHOP Exchange and purchase a Platinum PPO for its employees — the “UPMC Small Business Advantage Platinum PPO $250 $10/$25 – Premium Network” for which the cost of adding Ada would be $353.94 per month or $4,247 per year. For that Ada would receive a plan that had a unified $250 deductible and a unified $1,250 out-of-pocket limit.
But what if Babbage instead provided Ada with $4,247 in additional compensation? Ada could go into the FFM for Pennsylvania and find a policy similar to the UPMC one listed above. She could get the “UPMC Advantage Platinum $250/$20 – Premium Network” for a gross premium of $391.75 per month ($4,701 per year )with a unified $250 deductible and a $1,500 out-of-pocket limit. Ada would have to dig into her own pocket to do so, however. Because Ada is in likely in the 25% marginal tax bracket, she will net only $3,185.25 from the additional compensation. And, because Ada’s income is well above 400% of the federal poverty level, she will receive no advance premium tax credit to help pay for the policy. Thus, Babbage’s decision not to purchase on the SHOP exchange will end up costing Ada about $1,516 if she wants a comparable policy.
Ada is worse off largely because she gets no subsidy and because, by receiving cash instead of health insurance, she loses the current tax advantage offered by the latter. Thus, it is the wealthy individual who is hurt by having to go to the individual market.
All of the above is hardly a proof. There are lots more scenarios to be considered. But my initial conclusion is that the SHOP exchanges tend to be most useful only for the wealthy who would not get a premium tax credit were they to go to the individual exchange. Also since my preliminary research indicates that, on balance, SHOP policies are at least as expensive as individual policies and because the employer purchasing a SHOP policy generally ends the ability of the employee to get a subsidized policy on the individual exchange, the option to purchase in fact may hurt employees and their families.
So, we have to ask.: If we are going to keep some version of Obamacare, why not help stabilize the individual exchange pools by bringing some additional people into them: the generally healthy people who work for small employers.? So long as the market works and those employers pass on to their employees what they would have spent on the SHOP exchange for health care, most except the wealthiest are likely to be better off. Although one answer to this provisional suggestion is that doing so will hurt the federal budget — more people will get section 36B tax subsidies — there may be many who share the preference for a simpler Obamacare, and one that helps breaks the peculiar stranglehold that employer provided health insurance has had on this nation since the government distorted the market after World War II.
Amidst all the passion yesterday at the roasting yesterday by the House Oversight Committee of Glib MIT Professor Jonathan Gruber and Marilyn Tavenner, Administrator for the Center of Medicare and Medicaid Services, many have missed what may be the most important development of the day: Congress is closer to stopping the Obama administration from funding the Risk Corridors programs that insulates insurance carriers selling policies on the Exchanges from much of the financial risk. Chapter G of the Continuing Resolution currently in the works (the “Consolidated and Further Continuing Appropriations Act, 2015″) appears to block the Obama administration’s apparent plan of using a “slush fund” — the “CMS Program Management Account” — to pay insurers when obligations under the program exceed receipts. Many, including the non-partisan Congressional Research Service and Senator Jeff Sessions, believe that the earlier contemplated use of this account to pay for Risk Corridors was unlawful under the Antideficiency Act and Article I, section 9, clause 7 of the United States Constitution (“No Money shall be drawn from the Treasury, but in Consequence of Appropriations made by Law”).
The inability of the Obama administration to finance the Risk Corridors program is a direct threat to the operation of the Affordable Care Act. Insurers who priced policies based on the assumption that, if they went too low in their premiums, they would be protected against substantial financial risk by Risk Corridor payments from the federal government, will now be facing — to their surprise — an environment in which at least some of the Risk Corridor payments will not be forthcoming. Insurers contemplating entry or continued participation into the insurance markets created by Obamacare will now hesitate for at least 2016 — either that or they will price their policies higher to protect against the now assumed risk of loss. The effect for 2015 policies is unclear. In light of the forthcoming Supreme Court decision in King v. Burwell, insurers negotiated for a provision in their contract that gives them the ability to terminate their participate in the program if either cost sharing reduction payments or premium tax credits are not available to purchasers. They are not known, however, to have negotiated for a similar provision with respect to Risk Corridor underfunding and thus might be held by a court to have assumed that risk.
How severe the effect of this Risk Corridor limitation will be depends on how CMS uses whatever authority remains to make at least partial payments to insurers and, of course, the amount by which obligations under the program exceed receipts. Suppose, for example, that obligations to losing insurers under the Risk Corridors program are three times receipts from winning insurers. This means that losing insurers would receive only 33 cents on the dollar, at least until any future surplus from the program could make them whole. Such a result would likely infuriate insurers and induce them to seek further regulatory concessions from the Obama administration as a price of continued participation in the ACA exchanges. If as the Obama administration predicted, Risk Corridors will break even or even run a surplus, the limitation in Division G will have no effect at all.
In any event, the Continuing Resolution in which all this is contained is not yet law. And there are apparently many points of contention — some possibly even more important than Risk Corridors — up for debate. Who knows what weapons insurance lobbyists will bring to bear in the mean time to rid Division G of this critical limitation? If, however, Division G’s limitation on Risk Corridor payments survives, expect further trouble in the market for individual and small business insurance created by the Affordable Care Act.
It didn’t take long for my prognostication in the last paragraph to bear out. Insurers are already in an uproar. As reported in The Hill just now:
“American budgets are already strained by healthcare costs, and this change will lead to higher premiums for consumers and make it more difficult to achieve affordability,” said Clare Krusing, a spokesperson for the America’s Health Insurance Plans.
Wow! There’s a lot to say about the hearing that the House Oversight Committee just concluded. Here are some bullet points that I will try to flesh out in the days ahead. I’m going to start not with MIT Professor Jonathan Gruber, which is important, but with something yet more important that was addressed at the hearing: the planned escape of the insurance industry in the event the Supreme Court rules in King v. Burwell that the Affordable Care Act does not permit the executive branch to provide tax subsidies to residents of the many states that have, to date, not established their own health care exchanges.
King v. Burwell and the Gruber hearing
Under questioning from Georgia Republican Doug Collins, Marilyn Tavenner, the Administrator of Medicare and Medicaid Services, also a witness at the hearing, essentially confirmed (look here at 2:10-5:55) earlier reports that insurers had negotiated a provision letting them stop provide health insurance in the Exchanges if the Supreme Court were to rule that the federal government could not provide subsidies. The Obama administration evidently could not get insurers to participate simply by leaving the matter silent and telling the insurers they could rely on whatever protections the classical legal doctrine of impracticability would provide for “unforseen” circumstances.
Administrator Tavenner’s admission and the antecedent concession is is evidence that the insurance industry is very worried about the outcome of the Supreme Court decision. It is evidence that, despite its public disparagement of the lawsuit, the Obama administration understands that it represents perhaps the most serious challenge to the ACA in that it lets Justice Roberts and the Supreme Court purport to simply play umpire and leave to a theoretically functional Congress the task of fixing the statute. It is evidence of the chaos that is going to erupt when the Supreme Court rules that the Obama administration can not lawfully pay subsidies to individuals in the many states, including my home state of Texas, that have declined, to date, to establish their own exchange.
It’s also worth noting that some Republicans at the hearing focused on the absence of any appropriation by Congress for the cost sharing reductions that the ACA provides poorer purchases of Exchange plans. This absence has been one of the focuses of the lawsuit filed by the House of Representatives against the Obama administration. Secretary Tavenner purported not to know how much had been spent on this program. Expect this failure to appropriate to be a major wedge used by the Republicans in the continuing knife fight over Obamacare.
Gruber: Part 1
And now onto Professor Gruber. I fear his adventures in Congress are just beginning. He bizarrely declined to provide the Committee with the amount of money he received from federal and state sources for his work on Obamacare and its implementation, saying (1) that he didn’t know what the law required him to provide and (2) that, even apparently within $100,000, he couldn’t say how much it was. But Professor Gruber’s apparent disdain of money aside — ask me within $100,000 to tell me what I made on a consulting contract and I think I could manage it — his response begs the question of why he would not provide the information even if it were not required. What principle is his silence about payments protecting? Surely it is within the right of Congress to find out how much the federal treasury paid an individual, either directly or through grants to the states. It goes to whether legislative appropriations are being used properly and to the bias of the witness on other matters. So, expect a subpoena of Professor Gruber and quite possibly a return appearance. Also, I have a hard time understanding why Professor Gruber or his counsel thought it smart to prevaricate on this matter. I sure hope the MIT professor reported his income and paid his taxes on the grants.
Someone needs to really audit Professor Gruber’s modeling. If Professor Gruber believes that the ends justified the means with respect to the Affordable Care Act, that it was acceptable to articulate the plan in contorted ways and to distort its accounting in order to get it passed so that the American people would reap its benefits, why would that instrumentalism cease when it came to modeling? As someone who does economic modeling, I know, is part science, but it is also part art. There are choices to be made. Simply from a casual reading of the limited material Professor Gruber has released on his “GMSIM” model, it is apparent that he very much made choices. He selected various critical parameters in ways that were not approved of by other scholars. Did he do so as part of a genuine belief that others were wrong or because he needed those parameters in order to make the results come out “correctly?” Were there draft runs of the model in which the answers did not come out as he wished and were parameters tweaked in response. Until we see Professor Gruber’s code, until we see his drafts, we don’t know.
Perhaps with respect to Obamacare, this is all water under the bridge and a bit of political theater from the Republicans. After all, Obamacare, like it or not, is the law. But there is a philosophical issue involved and that is the vulnerability of Congress to laws predicated on modeling that is not validated and that may come from people who have their own political axes to grind or who are dependent on politically motivated sources of funding. Oh, and don’t be surprised if executive privilege is claimed to try and protect such potentially explosive documents.
The Democratic Response
The Democratic response to the hearing was particularly ironic. Yes, they were critical of Professor Gruber because he had — and this is surely true — handed the Republicans a PR coup gift wrapped with a ribbon. But, essentially the Democratic response was to tout the virtues of the Affordable Care Act. They did this throughout the hearing by reciting the many people that it had helped (although apparently not the husband of Congresswoman Lummis), by unrelenting claims of causation between the “bending of the cost curve” and Obamacare, and by putting on as their only witness an area man who, despite a pre-existing condition and likely high medical expenses from a “common medical condition,” could sincerely extoll the delights of managing to get less expensive insurance as a result of the Affordable Care Act. He also appears to have gotten better claims service, though it is hardly clear that this was the result of the ACA rather than the fact that his insurance policy now actually provided him with coverage for medical expenses sought to be reimbursed.
But what is the relevance of this Ari Goldman’s testimony other than to show that the ACA has helped him, in part by having someone other than the insured himself pay for his predictable high medical expenses and that the ACA has largely converted an insurance system based on risk into one that serves as a tool for implementing federal redistributionist policies? What is the relevance of assertions that Obamacare has slowed the growth in medical expenses — perhaps without a correlative decline in medical utilization? The only relevance is the instrumental argument that Professor Gruber made many times on the videos and recanted — but only insofar as it was articulated glibly — before Congress. The ACA is a good thing.
Perhaps the ACA is, on balance, a good thing. I depart from some on both sides of the political spectrum by believing that reasonable people can disagree on that issue both in the abstract and within the context of contemporary American political realities. But does that mean that it is acceptable to lie to people in order to get it past? Does that mean it is acceptable to get it passed by exploiting the economic ignorance of the American people by disguising taxes as non-taxes, by gaming the CBO scoring system, or by including programs such as the CLASS Act that were absolutely destined to fail but that could, thanks to the failure of the CBO to use basic principles of insurance accounting, yield a phantom $70 billion? We should not be so hypocritical as to believe that there are not many who have succumbed, at various times and for various reasons, to this rationalization. Professor Gruber had the poor sense to make that choice apparent and to attribute it to those in Washington. But to distance oneself from Professor Gruber’s statements but then defend his grilling by exclaiming the virtues of the ACA is essentially to recapitulate the very instrumentalism for which Professor Gruber was called on the carpet.
1. Best question of the day from Representative Cynthia Lummis comes in the context of Professor Gruber’s repeated claim that he was merely an economics expert, not a politican and his excuse of his prior statements by claiming that they were “glib.”
Lummis: “How many non-politicians know what CBO is? How many non-politicians know what scoring is? How many non-politicians would know that you have to get by CBO scoring in order to get the Affordable Care Act to say that it’s going to lower costs? You are a politician. Everything that has led up to your testimony today is inconsistent with your testimony today, which is to say all of your prior statements were a lie. Is that true? Were all of your prior statements a lie? Or were they just glib?”
I’ve been doing a lot of research on the state of policies sold on the health insurance exchanges. It’s not easy because the Obama administration, as even its friends acknowledge, has not been forthcoming with information. It has, however, placed some useful information in the public domain: two large databases of the plans being sold on the “Federally Facilitated Marketplace.” That’s the health insurance Exchange for states, like Texas and many others, that declined to establish their own exchanges. With the help of the R computer language, I’ve been sorting through this database and have reached the following conclusions.
The change in premiums between 2014 and 2015 depends significantly on the metal level of the plan and whether it is a PPO or HMO.
Gross premiums for platinum plans are up significantly in price, 21%, whereas bronze plans and catastrophic plans are down over 11%. The high increase in gross premiums for platinum plans creates a serious potential for an adverse selection death spiral in that segment of the market.
Net premiums will show larger percentage increases and decreases than gross premiums for many individuals. This is so because substantial parts of the premiums are paid via subsidies from the federal government.
PPO plans are up substantially in price, 8%, whereas HMO plans are down substantially, -18.%.
The combination of increases in the more generous platinum and PPO plans and decreases in the less generous bronze and HMO plans may start to divert Americans into healthcare plans that offer lower benefits and somewhat less choice, albeit at a lower price than was paid this past year.
Among plans that persisted between 2014 and 2015, the premium variations are less extreme: persistent bronze plans increased in price by 9.5% whereas persistent platinum plans increased in price by 14%. The larger variation in gross premiums overall is thus likely due to the exit of carriers who priced at extremes and low pricing by new entrants for bronze plans but very high prices for the more generous plans.
Cost sharing for the plans has increased somewhat, but many cost sharing arrangements have remained largely the same.
Competition, as measured by the number of unique issuers offering plans in each county, has increased substantially since 2014, but a market in which three or more insurers are actively competing is still a rarity, particularly for the plans that give consumers a greater amount of choice in selecting their doctor.
You can read a copy of the full report, which contains 29 tables of information, here.
As some of you have noted, there have not been any posts for about six months. I guess I needed a break and to focus on my day job. But the urge to blog on the ACA has returned and I will figure out some way to make time. So, stay tuned. I may not publish as frequently as I used to, but I will try to provide an independent perspective on this law.
The Congressional Budget Office issued a report this week revising its February projections of the cost of the Affordable Care Act. Although there is much to discuss regarding the report, I want to focus here on its troubling discussion of “Risk Corridors.” That’s the part of the law under which the federal government reimburses insurers selling policies on the new Exchanges for sizable fractions of their losses. It also taxes insurers if they happen to make money selling policies on the new Exchanges. Between February and April, the CBO estimated cost of Risk Corridors jumped $8 billion. In February, Risk Corridors were predicted to make the government a net of $8 billion over the three years of the program. Now, Risk Corridors are expected to net the government nothing. The CBO claims that this jump was caused by regulations issued by the Obama administration in March that drove up the cost of the program.
There’s a second explanation, however, for the $8 billion change between February and April that’s possibly more troubling. This past February I wrote a blog entry with a lot of math explaining that the CBO prior analysis of the Risk Corridors provision was baffling and rested on extremely dubious and factually unsupported assumptions about the profitability of insurers selling on the Exchanges. That error, if it was one, was particularly salient because it ended up forestalling growing efforts within Congress to repeal Risk Corridors as an unwarranted “bailout” of the insurance industry. Could it be that with the repeal threat gone, CBO is now using the “noise” created by an Obamacare regulation as cover for rectifying the unduly optimistic assumptions it made back in February regarding Risk Corridors? That would be very troubling, because while math errors merely challenge the CBO’s competence, the alternative behavior about which I am speculating here goes to something more important: the CBO’s integrity.
The CBO explanation means the Obama administration shoveled $8 billion to insurers through a regulatory “tweak”
The official explanation from the CBO on its change of $8 billion in the cost of Risk Corridors is as follows:
“In March 2014, the Department of Health and Human Services issued a final regulation stating that its implementation of the risk corridor program will result in equal payments to and from the government, and thus will have no net budgetary effect. CBO believes that the Administration has sufficient flexibility to ensure that payments to insurers will approximately equal payments from insurers to the federal government and thus that the program will have no net budgetary effect over the three years of its operation. (Previously, CBO had estimated that the risk corridor program would yield net budgetary savings of $8 billion).”
So, if the CBO is to be believed, the change isn’t due to any earlier error, but due to an administration regulation promulgated by the Obama administration that has resulted in a net of $8 billion more going to insurers. That’s a big change for several reasons. First, it means that the regulatory changes instituted by the Obama administration cost the federal government $8 billion. All of that money went to the insurance industry. And so, in March of 2014, without much fanfare, the Obama administration would in effect have written a check to the insurance industry for $8 billion. That payment would only have been motivated by one thing: a desire to keep insurers pacified and in the Exchanges after having deprived them of perhaps their most healthy potential insureds by a prior administrative ruling — in violation of the ACA — that insurers could keep selling non-compliant policies. The $8 billion would thus have been “damages” paid by the taxpayer in order to permit the President to honor his campaign promise that if you liked your insurance plan you could keep it.
In short, if you believe the CBO, a regulation for which statutory support will be extremely hard to find, resulted in the government shoveling $8 billion to insurers, basically to pacify them for the losses they suffered as a result of further regulatory changes of dubious legality. The Obama administration can not afford to have its signature program enter a death spiral as a result of regulatory actions that, while mollifying those who otherwise would have lost their health insurance coverage, caused insurers to lose more money in the Exchanges. And, again, the Obama administration did so in a clever way that made it difficult for anyone to have legal standing to challenge them. So far as I can discern, no insurer will be worse off as a result of the March 2014 regulatory changes. The real victims are taxpayers with diffuse interests and, of course, the Rule of Law.
The CBO math is still baffling
A second reason the change by the CBO is big comes from a look at the math. As I said in my February 2014 post calling the CBO February report “baffling,” consider the implications of asserting that the insurers would make so much money on the Exchanges that they would, on net owe the federal government $8 billion. If you do the math, it means that the CBO assumed that, over the course of three years, insurers would be earning about 8 cents on every dollar they earned via policies sold on the Exchanges. I just ran the numbers again and came up with a very similar conclusion: the earlier estimate could only be true if insurers were supposed to make a hefty 8% or greater return on premiums. That estimate of 8 cents on the dollar was really peculiar at the time because enrollment — let alone actually paying customers — was running seriously behind projections and the number of “young invincibles” was particularly low. Low overall insurance purchases and particularly low rates of purchases by the people who were most needed in the Exchanges caused many people to believe back in February that insurers would hardly make hefty profits and pay money to the government under Risk Corridors. Instead, they thought insurers would fare poorly and probably have to be subsidized (or “bailed out”) by the government.
The effect of the February CBO pronouncement was to dampen enthusiasm for a bill proposed by Senator Marco Rubio that would have repealed the Risk Corridors provision as a bailout to the insurance industry. If, after all, the federal government was, on balance, making money on Risk Corridors, it was hard to see it as a “bailout” to the insurance industry. Whether intended or not, the political effect of the February CBO announcement was to pull the rug out from one justification for repeal of Risk Corridors.
But is it even plausible to believe that the regulatory change made by the Obama administration in March without the approval of Congress could cause such a large change in the Risk Corridors program? I have done the math again and the answer is no. I do not see how it is possible to get $8 billion out of the regulatory tweak that was made. Again, the calculations are baffling.
Here’s how we know. The $8 billion the CBO thought back in February the government would make off of Risk Corridors represents about 4% of the premiums insurers on the Exchanges would take in during that time period. One can use that and other information from the CBO to reverse out a distribution for “allowable costs” (basically claims expenses) We can thus make a respectable estimate of how many insurers would make money on the Exchanges, how many would lose, and how much these insurers would gain and lose. I describe the gory process in my post from February. Call this distribution the CBO Insurer Profitability Distribution. Then assume the government tweaks, as it did, two regulatory parameters used in the computation of Risk Corridor payments, changing something called a profit margin floor from 0.03 to 0.05 and changing an “administrative cost cap” from 0.2 to 0.22. If one then takes the CBO Insurer Profitability Distribution and computes how much the government would now make on Risk Corridors does one emerge with the CBO’s new prediction that Risk Corridors will produce no net revenue? No! One gets that the Risk Corridors program now generates about 2.8% of premiums for the government. In other words, the reduction in Risk Corridor revenue resulting from the administrative tweak is only 1/3 of what the government claims.
The easier way to reach the CBO’s April’s conclusion is to assume that the gain of $8 billion resulted from two phenomena: (1) the regulatory tweak mentioned by the CBO and discussed above, but (2) a recognition that the CBO Insurer Probability Distribution the CBO had used in February was, as I have said, wrong. If, for example, one assumes that insurer claims were about 6% higher than the CBO estimated in February, the regulatory tweaks combined with higher insurer claims expenses indeed generate an $8 billion shift in the amount of revenue the government would make on Risk Corridors.
For those interested in the details, I link here to a Mathematica notebook showing the computations; I try to avoid black boxes.
So, what are we to make of this apparent discrepancy between the CBO’s explanation of its change in estimates and the actual effects of the regulatory changes it asserts to be the cause ? It could, I suppose, be my mistake. I have been careful and consider myself pretty knowledgable in this area, but I will hardly claim to be mathematically infallible. The problem is that the for ordinary Americans (like me), the CBO is a black box. It is not subject to the Freedom of Information Act and it does not publish enough of its methodology for even experts in the field to figure out what it is doing. That, I would submit, is a real problem for the democratic process, where the fate of legislation depends essentially on trust rather than the Reagan doctrine of “trust but verify” (doveryai no proveryai, in the original Russian).
It could also, however, be a coverup for a mistake (or worse) back in February. There is, after all, an alternative explanation of the change in estimate. It was unrealistic all along for the CBO to think that insurers in the Exchange were going to make money on balance. That’s what I suggested in my February 2014 post. So, rather than admit that the it had been guilty of unwarranted optimism, the CBO simply used a new distribution of likely claims expenses, came up with a different answer, and used the March 2014 regulatory changes as a smokescreen.
I will confess, however, that I am very uncomfortable with conspiracy theories or with theories that are premised on people acting in bad faith. Nonetheless, I would not find it impossible to believe that a culture could emerge in a politically sensitive agency that was reluctant to expose forcefully the consequences of government programs that proved far more expensive and far less successful than forecast originally. It would be a culture in which good news, or optimistic speculation, was uncritically embraced. What I challenge the CBO to do, therefore, not only with the Risk Corridors analysis, which is but the tip of a very big iceberg, but with the entirety of its ACA analysis, is to open it up for scrutiny. When government policy is essentially set on the basis of models that are not subject to peer review or public scrutiny, there is a great chance for error and, frankly, for manipulation. Government by black box breeds suspicion.
Postscript: Is the tweak legal?
I have said before and I say again that the regulatory tweak that the CBO now says will cost the federal government $8 billion is extremely dubious. It’s an extremely sneaky way of sending money to the insurance industry, resting, as it does, on arcane manipulations of mathematical formulae. And I have serious doubts that the changes are authorized by Congress.The submission of the original regulations in March, 2013 says that essentially all commenters agreed that a 3% margin for profit was appropriate. No commenters indicated at that time that insurers were entitled to a higher imputed rate of return on capital. No one said anything about 5%. Back in March of 2013, HHS thought 3% was the right number. There has been no fundamental change in the capital markets since that time. The only thing that has changed is that the Obama administration has made the pool of insureds making purchases in the Exchanges less healthy on average. The regulatory “tweak” moving the profit margin from 3% to 5% is thus not consistent with the original goal of Congress for the Risk Corridors program, but is simply a way of compensating insurers for another regulatory change.
The change in the administrative cost cap from 20% to 22% that will likewise result in higher payments to insurers is likewise dubious. The reason 20% was suggested in the original July 2011 proposal and chosen in the March 2013 regulations was to maintain parity with regulations governing the “Medical Loss Ratio” codified at 42 U.S.C. § 300gg-18 as part of the ACA. The idea, which was apparently supported by commenters on the original rules, was that if insurers — even small group and individual insurers — could not claim more than 20% administrative costs without owing rebates pursuant to section 10101(f) of the ACA then they should not be able to claim more than 20% administrative costs under the Risk Corridors provision. Makes sense! But, again, nothing has changed. There is no indication that anything President Obama did that raised the administrative costs of running a health insurance plan on the Exchanges. There is no indication that any factor in the real world (such as the cost of computers or paper) increased the administrative costs of running a health insurance plan on the Exchanges. The limits for the Medical Loss Ratio computation have not changed. There is no better reason now then there was a year ago to let the cap on administrative costs be higher for Risk Corridors than it is for Medical Loss Ratio. And, yet, it is now 22% instead of 20%. The only reason it has changed is to provide a vehicle for shoveling money to insurers.
Again, unless one thinks that the goal of keeping insurers in the Exchange is so overwhelming as to permit the Executive Branch to do anything, it is difficult to see a conventional, lawful justification for the regulatory change that results, according to the CBO, in $8 billion of compensation to the insurance industry. And I say that believing fully well that many of the Obama administration’s other regulatory changes — also of dubious legality — such as expanding the hardship exemption and permitting insurers to sell policies off the Exchanges that contain prohibited provisions have significantly hurt insurers selling policies on the Exchanges. Two wrongs do not make a right.
The following graphic shows the relationship between the Risk Corridor Ratio and the net receipts of the government for each premium dollar. As one can see, the higher the Risk Corridors Ratio, the less money the government receives or, in some instances, the more money the government pays out.
The following graphic compares the relationship between claims costs (“allowable costs”) incurred by an insurer as a percentage of premiums and the Risk Corridors Ratio. It does so for two sets of regulatory parameters. It first uses the regulatory parameters that were in place prior to March of 2014 (3% profit margin and 20% administrative cost cap). It next using the new regulatory parameters (5% profit margin and 22% administrative cost cap). As one can see, the regulatory changes increased the Risk Corridors Ratio for all levels of allowed costs and thus decreased the amount the government would receive from insurers (or increased the amount the government would pay to insurers).
Much has been made of the enrollment data in the Exchanges — pronounced “a success” by the Obama administration — and of the demographics of enrollments in the Exchanges — distributed enough towards the older end to be troubling. We don’t yet have the key data, however: what are the actual claims being filed by those newly insured on the Exchanges relative to what was expected. If the medical claims are higher than expected, that is likely to raise the costs of the Affordable Care Act this year and in the future. This year, high claims will increase “Risk Corridor payments” made by the federal government to losing insurers. Next year. it will place pressure on gross premiums charged by insurers or possibly force some insurers to withdraw from the marketplace. The result will be increased amounts paid by insureds, increased premium tax credit payments made by the government, and increased payments made by the government to address cost sharing reductions. It will also shrink the number of insureds below what it would have been had claims costs and, derivatively, premiums remained as expected.
Some evidence on claims costs is beginning to trickle in, however. Express Scripts, a large Pharmacy Benefit Management company that, so far as I know, has no axe to grind either in favor or opposition to the ACA, has published a report indicating that, at least so far, costs per member on the Exchanges are 35% higher than they are for commercial policies off the Exchanges. The study is based on a national sample of more than 650,000 pharmacy claims (423,000 covered lives) for the first two months of 2014 for patients enrolled in an Exchange policy with with pharmacy benefit coverage administered by Express Scripts. The analysis compared these pharmacy claims to those from commercial health plans, with pharmacy coverage administered by Express Scripts, during the same time period.
The key Express Scripts result — a 35% increase in claims costs — is significant for two reasons. According to data from the government’s own Actuarial Value Calculator, pharmaceutical expenses comprise about 21% of total healthcare expenses. Having to pay 35% more for such expenses is thus significant in and of itself. But peer-reviewed scholarly research such as that summarized and extended here indicates that pharmaceutical claims correlate positively with overall healthcare expenses. The higher pharmaceutical claims may just be the tip of the iceberg. Although these medical claims are often slower to be processed, Express Scripts has provided a disturbing leading indicator.
Before anyone pushes the panic button, however, there are less distressing possibilities. Everyone expected that those without prior health insurance or with lousy prior health insurance would result in a surge of claims to insurers for previously untreated conditions. One hopes insurers anticipated this surge in their pricing. If the surge is only transient as patients get various conditions under control, unanticipated extra costs on insurers will be addressed through Risk Corridor payments for this year and will not result in insurers revising their actuarial models of the risks posed by insuring on the Exchanges in an environment where most conventional underwriting methods are prohibited.
And, if one looks at the conditions that are apparently contributing to the high use of pharmaceuticals, a different spin can be placed on matters. The Express Scripts reports a higher use of anti-HIV/AIDS drugs, including expensive ones such as Atripla, among the Exchange population than among commercial insurers. So, it may be that the existence of subsidized coverage in the Exchanges is proving a vehicle for bringing hope and treatment to individuals with HIV or AIDS who previously were falling through the cracks (or being treated by other programs). The counter-spin, however, is that the fact that HIV treatment is a worthy end may be served does not mean that a general insurance scheme is the right way to address untreated HIV or AIDS. A coarsely rated and somewhat voluntary insurance scheme is a problematic vehicle for providing access to care to groups that include high-expense individuals, such as many with HIV or AIDS. The high and disproportionate prevalence of high expense individuals in a common pool contributes to the risk that the system will enter an adverse selection death spiral.
So, let me sound a notion of caution here. Just as the “enrollment” of 7 million people is not grounds for proclaiming the ACA a “success” or here to stay in any sort of stable way, an early datum about one component of claims is not grounds to proclaim the ACA a failure or to say with certainty that ACA insurance policies are likely to undergo massive premium increases. Still, since insurers will likely need to be making pricing decisions for 2015 in the coming months, and not after all the data is in early data can be important. The latest information from Express Scripts should be worrisome indeed.
Exploring the likely implosion of the Affordable Care Act