Insurance Reform Goes Crazy
While the National Association of Insurance Commissioners (NAIC) was rolling out a set of model laws for the states to consider in reforming their small group insurance laws, a handful of states decided to go much, much, further.
The NAIC models were focused on keeping the market working while ensuring access to coverage and preventing excessive rating variations between groups. It also proposed two alternative forms of risk sharing between carriers so no one insurance company would be hit too hard by a surge of high-risk enrollees. Almost all of the states adopted some variation of the rating restrictions but none adopted the risk-sharing provisions. A few years later the federal government adopted its own version of these requirements by enacting the Health Insurance Portability and Accountability Act of 1996 (HIPAA). These laws would go on to severely damage the market for small group coverage throughout the United States. We’ll come back to that in a future post.
Meanwhile, a few states went well beyond these requirements and decided to apply even more onerous restrictions to both the individual (non-group) and small employer markets. New Jersey, New York, Maine, Massachusetts, Vermont, Connecticut, Washington, and Oregon all applied “community rating” to one or both market segments, along with “guaranteed issue” of some form.
Community rating means that all enrollees are charged the same premium regardless of their risk factors, and guaranteed issue means that all applicants must be accepted for coverage whenever they apply. In some cases the states also adopted a limited number of allowable standardized benefit plans.
Economists of every political persuasion warned these states that such restrictions cannot work in a voluntary market. At a minimum they will discourage the young and healthy from enrolling until they know they will need services, which will spur a “death spiral” of decreasing enrollment and ever-higher premiums. Liberal economists concluded that enrollment should therefore be mandatory, while conservative economists concluded that the restrictions should not be adopted at all. But both sides agreed that combining voluntary enrollment with these restrictions would doom the market.
But most of the politicians and “consumer advocates” in these states had no interest in listening. They were in such a fever to “do something” that reason was thrown out the window. It may also be possible that some of them did understand the consequences and knowingly adopted the restrictions anyway, thinking that the sooner the private market was destroyed, the sooner they would realize their real dream of a government-run health care system.
The late Conrad Meier of the Heartland Institute did a nice job of summarizing the experience of all eight states in a 2005 publication, “Destroying Insurance Markets.” This publication describes the precise experience of each state. He quotes the Council for Affordable Health Insurance (CAHI) from 1993 as warning against this approach to insurance regulation:
Most of today’s uninsured are young and do not have much money. Community rating forces them to subsidize the cost of the middle-aged, who are at their peak earning power. Forcing the young to pay more will drive them out of the insurance market, raising costs for everyone.
Meier went on to look at what had happened ten years later and found that:
- Between1994 and 2003, the share of the population in these eight guaranteed issue states covered by individual health insurance plans fell dramatically.
- The eight states have seen a massive exodus of private insurance companies that had been selling individual health insurance policies.
- Premiums for individual insurance have soared.
- By contrast, states that did not adopt guaranteed issue and/or community rating have seen much smaller premium increases.
Meier took a closer look at each state, but we will confine our examination here to New Jersey, the poster child of these changes. He focused on three issues: raising premiums, declining coverage, and abandonment of the state by carriers.
Ten years after the reforms went into effect the rising cost of coverage was jaw dropping. For instance, Aetna’s Plan “D” family premium went from $769/mo. in 1992 to $6,025/mo. in 2005. The increase for the New Jersey Blues was similar. These are neither the worst nor the best examples. Plan “D” is a fairly rich benefit design, but not outrageously so, with a $500 deductible and 80/20 coinsurance. Meier reported that in 2005 the lowest monthly premium for single coverage for each of the five plan designs was:
| PLAN |
DEDUCTIBLE |
CARRIER |
MONTHLY PREMIUM |
| Plan A | $1,000 |
Oxford |
$517/mo. |
| Plan B | $1,000 |
Aetna |
$756/mo. |
| Plan C | $1,000 |
Oxford PPO |
$468/mo. |
| Plan D | $ 500 |
Oxford |
$1,371/mo. |
| HMO | $ 15 copay |
BCBS NJ |
$494/mo. |
Source: “Destroying Insurance Markets”
It is worth noting here that in states without community rating a 24-year old male would likely be charged under $100/mo. for coverage like this.
The number of people covered by individual coverage in New Jersey dropped dramatically, though the count varies considerably depending on who is doing the counting/ Meier reports that the U.S. Census Bureau counted 998,000 people with non-group coverage in 1994 and 623,000 in 2003. The Employee Benefits Research Institute put it at 500,000 in 1992 and 300,000 between 2001 and 2003, while the state on New Jersey said coverage fell from 156,565 in 1993 to 78,298 in 2003. In all three cases enrollment fell substantially after the reforms were enacted.
In an article in Health Affairs in 2004, Alan Monheit and colleaguesconfirm these assessments, writing that, “despite positive early evaluations, the IHCP (New Jersey’s individual market) appears to be heading for collapse.”
The “positive early evaluation” Monheit refers to is an article by Katherine Swartz and Deborah Garnick written in 1999 that was gushing in its praise for the program, calling it an “unprecedented achievement.” To be generous, perhaps Swartz looked at it too early, before the perverse effects had really kicked in. But for years later the “policy community” kept citing the Swartz paper as proof that New Jersey was working just fine and those of us who predicted otherwise were delusional.
But Swartz was wrong. Monheit writes:
The IHCP’s current situation points to a market that is heading for collapse. Enrollment has declined from a peak of 186,130 lives at the end of 1995 to 84,968 at the end of 2001. In addition, premiums have increased two- to threefold above their early levels. These changes have raised concerns as to whether a comprehensive regulatory effort such as the IHCP can yield a sustainable health insurance market.
The reasons are obvious to anyone not blinded by a political agenda:
Since pure community rating imposes the same premium on low- and high-risk people, the premiums of low risks exceed their actuarially fair level, while those of high risks are lower than their fair level. A sustainable market equilibrium may be tenuous under such a requirement.
Monheit also points out that the small group market was relatively healthy during this period, in part because New Jersey was enjoying economic growth (and job creation) and because the Small Employer Health Benefits Program, which was implemented about the same time, allowed for greater rate variation based on age and location. Indeed, while premiums in the individual market more than doubled from 1996 to 2000, small employer insurance grew only 30%. So enrollment in this segment grew from 694,312 in 2004 to 937,784 by the third quarter of 1999, before dropping again to 884,104 in the third quarter of 2001.
In the past few years New Jersey dropped its pure community rating requirement in the individual market, but one is left to wonder how many families were financially destroyed in the meantime by this misguided social experiment. How many people found they could no longer afford coverage and were thrown into the mercies of the charitable system? How many people continued to pay politically inflated premiums and were forced to curtail other important services?
No one ever bothers to count the people damaged by liberal social engineering. So people like Katherine Swartz continue to sleep well at night, heedless of the damage she has caused.
I hear numerous people talk about the need to broaden the risk pools to make premiums affordable. What they really mean is to force young people to pay premiums that far exceed their expected costs in order to subsidize older (read: wealthier) people who get a bargain. Greg is correct that community-rated premiums discriminates against the young, who make the rational decision to go “bare” and be uninsured.
John, Devon and you other smart people;
How much of the problems of individual/family policy availability could be remedied by eliminating from the federal tax code the bias for employer provided group policy?
If all employer provided insurance was treated as income to the employee and not a business deduction freebie, insurance companies would need to either get back into the individual market more or get out of the game all together. Employees would not want these all inclusive plans because of the price/income and in union shops the emphasis on excessive benefits would decrease.
Also, if group policy plans were to be organized would not opportunities open up outside the work place? Like balding men over 6 feet tall and their dependents? Policy ownership separate from the workplace would allow employees to sing that old Johnny Paycheck song, “Take This Job and Shove It!”.
Eliminating the tax code bias at the federal level does not ensure that insurers can accurately price risk within a state because, contrary to the Obama administration’s claims, individual insurance was regulated by the states until the federal government decided to make it one of its enumerated powers.
As Greg points out, by passing community rating these states ensured that risk couldn’t be accurately priced for each member. The result was that they destroyed large swathes of their individual markets.
Linda, did not the federal government create the state cartels by restricting interstate sales? Yes, states have made stupid political decisions, too.
The reason I focus on the federal tax code is the old saying about following the money. And thanks to FDR there has been a distortion in incentives in the medical insurance system and medical care costs for 60+ years.
Dr. Steve,
You are right. Because of the current tax code anyone who can will get employer-sponsored coverage. This itself is the first tier of selection since people who are employed are by definition able to work. That leaves the individual market only to those who are not regularly employed — too sick to work, partly retired, working irregularly, etc. That means the individual market is loaded with people who are high-risk either financially of in health.The individual market is sort of a “residual market” for employers.
This, in turn, means that writers of individual coverage need to be careful about who they accept.
I contend the individual market could work very well if it did not have to compete with the highly subsidized employer market.
Greg, my point exactly, but better state by a professional. Thanks.
Does the federal government restrict interstate insurance sales? I thought it was the states, who require that any product sold in them be licensed and run according to their regulations.
Not everyone in the individual market is ill or only employed part-time. I live in a state with a higher than average proportion of people who own their own business. A lot of them have individual policies. My state also had a fairly high number of HSA qualified individual policies as well. Its individual market worked pretty well until the local regulators got on the regulatory bandwagon.
Linda, I may be mistaken, but I think Commerce has allowed states to restrict interstate commerce in that market. I cant remember the details. That is one of the things advocated by some to open markets across state lines and the federal government has to change the rules.
This is a good post, Greg. I shall share it with my students.
Uwe
Employees, not employers, are the beneficiaries of the exclusion from taxable wages of health insurance premiums paid by employers. Employers have always been allowed to deduct compensation expenses, whatever their form. This is why better compensated people – people in higher tax brackets – will prefer an additional dollar of health “insurance” (pre-paid health care) to an additional dollar of gross pay. Gross pay will be hit with a 30 to 40% tax. Pre-paid health care will not.
I agree premiums paid by employers should be taxed. But while I think that is a necessary condition, I don’t think it sufficient. It will take having a robust and competitive individual health insurance market as well. Otherwise premiums paid through employers will likely still be lower because those with jobs are a healthier risk pool that excludes those who cannot work for health reasons. It will happen over time, of course, as younger, healthier employees discover less expensive other otherwise more desirable options in the individual market than their employers offer.
But for such options to even exist, the individual market must not be both community rated and guaranteed issue. And employees must be given the option to make the choice about whether to take the compensation as insurance or wages…
This is harder than it seems. Employers have several reasons to want to maintain their group policies.
Employer competition is one. Which employer in an industry is going to be the first to shift the burden of selecting a plan from the unfiltered marketplace to employees? This choice is real work… far harder than selecting car insurance.
Employers have an interest (albeit secondary to employees’ interests) in a healthy workforce. Health insurance actually protects the employer from avoidable disruptions in the workplace caused by workers who do not value their health enough to protect it. Will states allow health status to be directly considered as an employment criteria? Can employers prevent the states or the feds from making “healthfulness” a protected class, like race, creed, color, religion, sex, sexual preference, disability, etc.? I have my doubts.
Employers are run by people who tend to be both older and richer than their employees. Thus the benefits of “community rating and guaranteed issue” WITHIN the employer are more likely to accrue to the people running the company. Not to mention those whose sole job is managing benefits within a larger organization are unlikely to recommend firing themselves. While they could start new “health insurance consultancies” to help employees sort through their options, they are unlikely to prefer that over the certainty of their current employment.
So it’s far from sufficient to tax health benefits. It might provide a big boost to government tax receipts, and that tax would be disproportionately borne by higher income people (the value of the exclusion to a minimum wage person is far less than it is to someone making $100K/yr.)
From a purely economic perspective, the exclusion is responsible for some over-allocation of resources to health insurance, and through insurance, to health care. That’s why I’d like to see the exclusion eliminated. I’d like to see overall tax rates lowered a bit to prevent it from becoming a deadweight loss to workers today.
As long as you have multiple private insurers with segregated risk pools, the policy issues discussed will remain a problem. Guaranteed issue doesn’t work because of adverse selection, unless there is a mandate for everyone to be insured. But then community rating would be a problem because of the higher premiums that younger, healthier, but lower-income individuals must pay, as mentioned in this article.
Since adequate health care coverage is now too expensive for median-income families, some form of subsidization is required – either the employer contribution (which is actually paid by employees in forgone wage increases) or government subsidies as in the Affordable Care Act. Getting the subsidies right is almost impossible since there are too many variables to determine precisely how much of a subsidy each individual or family actually requires (which is partly why it is estimate that over 20 million would still be uninsured in spite of the mandate).
It would be far simpler and much more equitable to totally separate the financing from the delivery of care. Financing should be based on ability to pay, and health care services should be based on medical need. This can be accomplished quite easily by establishing a single universal risk pool, funded equitably through progressive taxes, and then providing everyone with the care that they need, when they need it.
This, of course, would be the “Improved Medicare for All” that some of us keep promoting. It really is a great idea.
Greg, are you sure you aren’t arguing this both ways? Either group coverage is expensive and nobody wants it (because of community rating) or it’s cheap and everyone wants it (because of the tax incentive). But it can’t be both for the same individual.
The tax incentive only goes so far. For a young, lower-wage employee, I can’t believe that the tax incentive is enough to compensate for the higher cost of group insurance. More likely, this employee (to the extent that he is aware of the situation) would try to opt out in favor of a higher taxable wage, or else look for a job that lacks coverage but pays more.
For most states, we seem to have backed into a system where modified community rating is the norm only if you have access to employer-sponsored coverage, and where a tax incentive takes the place of Obama’s individual mandate.
You may be able to make a case for abolishing employer-sponsored group plans and forcing everyone to shop for individual coverage, but good luck trying to dislodge the status quo. Even if I were sold on the idea, it wouldn’t be my top priority now that the real Obamacare is in the process of becoming established.
Richard Relph —
Excellent comment. You are right that (some) employers will continue to have an interest is providing coverage and there are some marketing efficiencies in doing so. But the tax code should not be tilted in that direction.
Don McCanne —
I think that is your strongest argument — income-based financing. The case breaks down when you want to rely on it exclusively and have the government pay for everything. What is covered then becomes a political decision. As long as 51% of the population wants it who cares about the other 49%? The biggest objection to income-based financing is where we stop. Is it “fair” that the poor pay the same for a loaf of bread as the rich? This destroys the whole purpose of prices.
Bart —
I don’t think I said group coverage is more expensive. There is no reason it should be absent the richer benefits of most group plans. You are right that people who pay no (or little) taxes get no benefits from the exclusion, but as Richard pointed out they aren’t the ones making decisions here. It is interesting that community rating within a group has exactly the effect we would expect. That is, young employees are twice as likely to opt out of employer plans as their older colleagues. Finally, I have no interest in abolishing employer group plans for all the reasons you cite. But I am all for leveling the playing field in tax law so people can make rational decisions on how best to obtain coverage.
Don, well spoken… just wrong. Here’s why.
Insulating people from the costs of health care is precisely how we got to the point where ordinary people cannot afford ordinary care!
“Medical need” is not like 2000 calories a day and 64 ounces of water. (Nor is ability to pay, though we’ll ignore that for the moment.) Medical need cannot always be objectively determined. Which leads to ALL KINDS of opportunites for inefficiencies and outright fraud to creep in… unintentional and otherwise. Ever wonder why there are TV commercials for electric wheel chairs? Or drugs? It ain’t objective “medical necessity”… If it were, advertising would be irrelevant.
Further, how do we reward the slightly better providers and punish the slightly worse if quality, timliness, satisfaction, availability, accessibility, etc. are all completely irrelevant to payment, which is based on a code number? (The best and the worst will be dealt with outside third-party payment… concierge for the best; torts for the worst.)
Equal payment for so-called “equal” work merely forces the better providers to seek their compensation through other channels. “I’ll take you on as a patient if you get my kid in to your private school, or give me a discount on my next car, or …” The poor will still have a disadvantage in the real marketplace, where individual doctors treat individual patients. It is pure delusion to think that all doctors performing XYZ procedure are equally good, which is the very premise of equal payment. Where legislators effectively outlaw monetary payment for marginal differences in quality, convenience, etc., other compensation mechanism WILL be created. There MUST BE some way to balance supply and demand at the micro level… between individual patients with their individual preferences and doctors with theirs. All the “universal system” designers simply ignore this reality because it is precisely this “knowledge problem” that central planners cannot solve.
When health care is “free”, or there is no or minimal marginal cost to consuming more at the micro level, demand will rise. Those third party payers – the taxpayers in your “system” – will try to limit the macro cost by cutting payments to providers. This will eventually lead to shortages of providers, and, as a natural consequence, longer waits. Sometimes fatal waits. This has been the story all over the world. It takes time to play out, of course, because doctors are both professionally and economically disinclined to stop practicing even as their incomes fall. But at the margins, it is occurring.
So what is an individual doctor – with limited time available – supposed to do when faced with more people wanting services IF they cannot raise prices??? They WILL – of necessity – find a mechanism to decide who “wins” and who “loses”. Maybe politcal connections? “In kind” services? Preferring people with certain attributes – beauty, intelligence, church affiliation, etc. Something the DOCTOR values that the patient can provide must be tendered in lieu of the marginal increase in price that ordinarily would balance supply and demand and signal opportunity for investment.
Until you can design a system where the people producing the demand are directly impacted by the cost of the supply – which is the EXACT ‘feature’ all the universal systems are designed to eliminate – you will be trying to design the perpetual motion machine. Something that looks good on paper, but can and will never work in practice.
Greg, I don’t see how you can abolish the tax differential between group and individual plans without rapidly eliminating the former. Employers may have incentive to continue to provide group coverage, but how would they cope with adverse selection? They might offer a taxable allowance and leave it at that. What else could they do?
I just don’t see this happening.
On the other hand, it should be possible to offer a tax credit for non-employer plans that are subject to the same rules that employer plans have to follow. In effect you would have non-employer group plans competing on a level playing field with the employer plans.
Hopefully the non-employer side of things would be set up more rationally, e.g. with smaller but more consistent tax breaks and more moderate rating restrictions. It may not be the direct route to the libertarian ideal you seem to favor, but at least it opens up a migration path in that direction.
Replacing Obamacare should be the priority; I don’t see how you do this without addressing at least some of the concerns of its supporters.
Devon – to F/U your comment on high risk pools I want to share a comment from a June 23, 2010 report on high risk pools by the Congressional Research Service. It indicated:
Supporters of high risk pools – – – reason that by removing high risk persons from the individual market and placing them in publicly subsidized insurance pools, coverage in the individual market will become more affordable. They argue that better risk spreading helps to stabilize the market, promote competition, and retain insurance carriers—earning the support of such organizations.
Here in Connecticut our risk pool does the same thing in the small group market. HIPAA requires guaranteed issue in the 2 to 50 market. Thus, companies, after reviewing the health history report the person submitted, can place an individual in the pool. That individual will not know this and their ID card and benefits are the same as other EEs in the firm.
Our small group market is not fully community rated. Ins. companies can adjust their rates based on factors such as zip code and age.
Our high risk pool has a guideline that premiums will be 150% of the market. One company provides the coverage for everyone in the pool. However, all insurance companies marketing in the state make up the difference between premiums and pool claims by paying a fee based on their percentage of covered individuals in the state.
Here’s an excellent article on Heath Care costs that shoulf be considered in this dbate:
http://economix.blogs.nytimes.com/2011/11/11/equalizing-payments-for-medical-care/?partner=rss&emc=rss
Uwe E. Reinhardt is an economics professor at Princeton. He has some financial interests in the health care field.
For some time now, health policy makers around the world and the analysts who advise them have been exploring reforms of the methods by which the providers of health care are paid or, as the latter prefer to call it, are “reimbursed” – an unfortunate, mind-altering expression on which I have already commented in an earlier post.
Today’s Economist
Perspectives from expert contributors.
Several papers in the current issue of Health Affairs are devoted to that topic. They include one by me, another by David Miller that addresses large variations in Medicare payments for surgery and one by Peter Hussey about bundled payments for treatments of an entire episode of illness. See link for Health Affairs – http://content.healthaffairs.org/content/30/11.toc
My paper focuses on the fact that every private health insurer in this country pays different physicians, hospitals and other providers of health care different prices for identical services. The flip side of this practice is that a given provider charges different payers – insurers or patients – different prices for identical health service. Economists call that practice “price discrimination.”
Figure 1 below illustrates this phenomenon from the insurer’s perspective for colonoscopies in New Jersey. Figure 2 shows payments in 2007 by one large private insurer for appendectomies (then code DRG 107) and coronary bypass grafts with cardiac catheterization (code CABG, then DRG 107) in California at what are known as “tertiary hospitals” — those with the ability to support medical specialists in medicine, pediatrics, obstetrics and gynecology, surgery, their subspecialties and ancillary services.
New Jersey Commission on Rationalizing Health Care Resources, Chapter 6
New Jersey Commission on Rationalizing Health Care Resources, Chapter 6
While there evidently is pervasive price discrimination within the private health-care sector, there are also sizable price differentials between public payers on the one hand and private payers on the other. Figure 3 illustrates this phenomenon by means of the so-called payment-to-cost ratio paid to hospitals by Medicare, Medicaid and private health insurers. This ratio is calculated as a fraction of the full costs that hospitals report to have incurred for patients covered by these three payers.
American Hospital Association, Annual Survey Data for Community Hospitals – Trendwatch Chartbook 2011, Table 4.4.
We can see that in many years, hospitals report that the two public payers have not covered the full cost (including presumably allocated fixed-overhead costs) of the care for the patients covered by these programs. Private payers, on the other hand, pay a sizable margin on top of the full costs of treating their covered clients.
Private health insurers and their principals, private employers, deplore this phenomenon as a “cost shift” from public payers to them. Of course, by similar reasoning, private insurers and individual self-paying patients who pay relatively high prices for given services can lament that they are the victims of a cost shift from private insurers with stronger market power vis ŕ vis hospitals, which enables them to bargain for lower prices for identical services.
Many economists do not buy into this contention, asserting that price discrimination can exist without a cost shift. Explaining how economists arrive at that conclusion goes beyond the limit of this post. Interested readers, however, might consult my previously cited paper in Health Affairs, or, if they are not daunted by more formal economic analysis, read Austin Frakt’s critique of the cost-shift theory.
Whether or not one accepts the cost-shift premise, the question arises of whether price discrimination in health care has served the United States well. On this point, the business school professors Michael Porter and Elizabeth Teisberg have this to say in their book “Redefining Health Care”:
This administrative complexity of dealing with multiple prices [for the same service] adds costs with no benefit. The dysfunctional competition that has been created by price discrimination far outweighs any short-term advantages individual system participants gain from it, even for those participants who currently enjoy the biggest discounts. The lesson is simple: skewed incentives motivate activities that push costs higher. All these incentives and distortions reinforce zero-sum competition and work against value creation.
Similarly, in commenting critically in The New England Journal of Medicine on another study of administrative costs of Canadian and American health care, Henry Aaron of the Brookings Institution offered this preamble to his critique:
I look at the U.S. health care system and see an administrative monstrosity, a truly bizarre mélange of thousands of payers with payment systems that differ for no socially beneficial reason, as well as staggeringly complex public systems with mind-boggling administered prices and other rules expressing distinctions that can only be regarded as weird.
Other nations with multiple insurance carriers – for example, Germany and Switzerland – avoid price discrimination in health care through negotiations over prices between regional associations of health insurers and counterpart associations of health care providers, subject to some overall budget constraint informed by macroeconomic conditions (e.g., growth of the payroll on which premiums are based or per capita income). The negotiated prices then apply uniformly to all insurers and all providers in a region (states in Germany and cantons in Switzerland). In the United States, Maryland has operated such an “all payer” system for hospitals for several decades.
An all payer system has the potential to reduce health care costs in several ways, including the discipline of a fee schedule on providers, the schedule to be negotiated with providers on a regional basis, and a reduction in the enormous complexity and high administrative costs of the current system.
If it is desired, an all payer system can also serve as a way to constrain the future growth of health care to some desired path – e.g., with total national health spending growing annually only 0.5 percentage points faster than the growth of the rest of gross the domestic product, rather than the traditional two percentage points faster that prevailed over the last four decades. A two percentage point differential is simply not sustainable.
In my above-cited paper, I advocate an all payer system for the United States to eliminate the pervasive price discrimination inherent in American health care and, to the extent it exists, the much-lamented cost shift. For readers of this blog, Health Affairs has graciously provided access to the paper until Nov. 16. I invite readers to take a look at my argument for an all-payer approach and share with us their reaction to it.
Ann Molison
Health Care for All Colorado
http://www.healthcareforallcolorado.org
Rick,
Your comment is pretty badly off-topic. You may want to weigh-in in John Goodman’s post on bundled payments at — http://healthblog.ncpathinktank.org/surprising-fact-of-the-day-there-is-no-such-thing-as-an-unbundled-medical-bill/