Fractal Geometry, Actuarial Risk, and §1332 Waivers—The Role of the States in Reforming Health Care

By: Alden Bianchi & Edward Lenz

September 25, 2017

At this writing, the prospects for success of the latest Republican effort to replace the Affordable Care Act appear bleak—but the Graham-Cassidy bill on which the GOP has pinned its last-ditch hopes highlights a major political and policy flashpoint in the fight to repeal, replace, or repair the law: the degree to which states should be free to innovate and experiment by adopting non-standard health insurance product designs in their individual and small group markets. Under current law, there is little flexibility. Proposals abound to change this, but to do so invites consequences with which lawmakers must be prepared to deal—involving complex economic and actuarial issues and fundamental questions regarding the role of the federal government and the states in health care.

This post addresses these issues.


The stakes underlying the efforts to reform the Affordable Care Act could not be higher: health care consumes about one-sixth of the U.S. economy, an amount that is materially higher than the health care expenditures in all of the other of the world’s developed countries. As the ill-fated efforts to “repeal and replace” the Affordable Care Act—e.g., the American Health Care Act (in the House of Representatives) and the Better Care Reconciliation Act (in the Senate)—demonstrate, reforming how the U.S. finances health care is no simple matter.

At the highest level, there is a stark partisan divide over who pays for the cost of medical care—and to what extent the federal government should undertake this obligation. In the case of Medicare and for much of the American workforce covered under employer-sponsored plans, there appears to be broad, though at times grudging, bipartisan agreement that favors the maintenance of the status quo as to who pays. Where Medicaid and the individual market are concerned, however, the opposite is true. The political right and the political left are deeply divided.

Complicating matters is that there appear to be two different challenges confronting policy-makers: controlling the costs of medical care, on the one hand, vs. constraining premium costs, on the other. Though related, these are not the same.

  • In the case of the costs of medical care, the goal is easy to state, if difficult to achieve: can government or the private sector take steps that “bend” the proverbial “cost curve,” i.e., bring the annual growth rate of the aggregate costs of medical care in the U.S. (4.9% for the period 2007 to 2014, according to the Kaiser Family Foundation) within the annual rate of real growth in the country’s Gross Domestic Product (generally around 3%). Some progress has made in this regard, although the contributing factors are difficult to assess.
  • In contrast, premium costs pose daunting political questions. Less generous coverage, or coverage with significant gaps can lower premiums, but at the same time leave individuals to self-fund the costs of care. While the possible exposure to some uninsured claims or portions of claims might not bother younger or healthy individuals, it is a grave concern to older individuals or those who have pre-existing conditions or are otherwise in less than optimal health.

While the efforts to repeal and replace the ACA pose many issues, at least three are getting the most attention: withdrawing from the ACA’s individual mandate, funding ACA cost-sharing amounts owed to carriers, and permitting states greater latitude in the design of the health insurance products. We deal here with the latter.

Under the ACA, insurance products sold in the individual and small group market are required to provide a comprehensive list of medical goods and services referred to essential health benefits. While states have some latitude to vary plan design under so-called “Sec. 1332 waivers” they are not free to entirely forego essential health benefits. Many of the ACA “repeal and replace” proposals adopt a more liberal view. They aim to give states far greater flexibility is the design of health insurance products offered for sale by their state-licensed carriers.

(For an explanation of the various efforts to repeal and replace the ACA, please see the recently issued ML Strategies Advisory entitled, Evolution of Repeal and Replace prepared by my colleagues, Rodney Whitlock, Kate Weider, and Eli Greenspan.)

What fractals have to do with health insurance plan design policy coverage?

In his compelling new book entitled, Scale: The Universal Laws of Growth, Innovation, Sustainability, and the Pace of Life in Organisms, Cities, Economies, and Companies, Geoffrey West traces, among other things, the relatively recent development of fractal geometry. A theoretical physicist, Dr. West applies the laws of fractals and scaling phenomenon in variety of contexts other than health insurance (a TED Talk by Dr. West on the subject of his book is available here). But his views of scaling phenomena tell us something about our current predicament as we ponder how to best regulate health insurance markets.

It is unnecessary to understand the complex mathematics of fractals to see the implications for health insurance. It is sufficient to note simply that the term “fractal” means and refers to a self-scaling phenomenon; that is to say, a twig looks similar to a branch; and a branch looks similar to a tree, but they are merely similar, not exact replicas of each other.

Extending this concept to insurance, one can observe that a properly structured insurance market covering the entire country (e.g., Medicare) distributes risk in a manner similar to, but not exactly like, a market covering a major corporation, each of which in turn distributes risk in a manner similar to, but not exactly like, the individual market within a state—the differences depending primarily on the particulars of each risk pool. Fractal geometry predicts insurance markets will operate similarly irrespective of size. But this assumes a properly structured market, and it is here that we need to consult actuarial science.

The role of actuarial risk

The concept of risk pooling goes to the heart of all forms of insurance. In the context of health insurance, the risk pool consists of the set or group of individuals whose medical costs are combined to calculate premiums. In a properly functioning risk pool, the higher costs of the less healthy are offset by the relatively lower costs of the healthy. The concept applies with equal force whether the pool covers employees and beneficiaries of a large or medium-sized self-funded employer, or the pool that is created under a state small group law or individual market. (For an excellent explanation of risk pooling in the individual health insurance market, please see “Risk Pooling: How Health Insurance In The Individual Market Works” by the American Academy of Actuaries, available here.)

A stable group is one that has not been adversely selected against. An insurance market or pool is subject to adverse selection where the group includes a disproportionate share of unhealthy individuals. This happens when individuals with greater health care needs are allowed to purchase coverage with more generous benefits than their healthier counterparts. One way to prevent adverse selection is to compel the participation by healthier individuals. This may be done through an individual mandate, as in the case of the ACA, or through a continuous coverage requirement, as in the case of the HIPAA portability rules. But if such steps prove ineffective and the pool becomes unbalanced due to adverse selection, costs could over time spiral out of control as premiums rise and healthier individuals drop or seek less expensive coverage.

ACA Section 1332

To encourage state innovation in health care coverage, Section 1332 of the ACA authorizes the Department of Health and Human Services to waive certain of the ACA’s requirements beginning in 2017. States may seek Section 1332 waivers of requirements related to the essential health benefits (EHBs) and related cost sharing limits. Also waivable are items such as premium tax credits and cost-sharing reductions. States may also modify the ACA individual and employer mandates under the Section 1332 waiver authority. They may not, however, sanction the sale of insurance with less than comprehensive coverage, and what coverage is made available must be otherwise affordable.

The proposed expansion or waivers under the ACA repeal and replace proposals

A key feature of the various proposals to repeal and replace the ACA is an expansion of the ability of the states to obtain waivers of certain Federal requirements. These generally include, among other things, allowing states to waive the ACA’s essential health benefit requirements. To take an extreme example, carriers could design a plan that covers only preventive services and aromatherapy. Perhaps a more realistic example might be a plan that covered inpatient and outpatient hospital and physician services but does not cover infertility or prescription drug coverage.  Preventive-services-only plans are currently available on a self-funded basis at a cost of about $60 per month. A plan that covers inpatient and outpatient hospital and physician services but not infertility or prescription drug coverage might cost on the order of $200 per month for self-only coverage.  This is in contrast to an average cost of a silver-level product on a state marketplace of about double that amount.

It is not difficult to spot the problem: younger, healthier individuals would be inexorably drawn to the $200 product, leaving older, less healthy individuals to purchase the more comprehensive and expensive silver-level (or higher) plan. The result is that the risk pools of these two products would not be “self-similar,”—i.e., they don’t scale in any mathematical sense. Rather the former would consist predominately of young and healthy individuals, while the latter would consist predominately of the older and sicker individuals.  Such a market is not necessarily unstable, of course (although it strikes us that there might be a point at which waiver authority is so broad so as to invite the approval and sale of a product with the capacity to destabilize a state’s small group or Individual market).

Recall that cost is a function of the average health care costs of the risk pool’s enrollees. So it is entirely possible that the insurance markets that arise under an expanded waiver regime would be stable and otherwise commercially viable, but it does mean that the older and sicker individuals may pay higher premiums when compared to current law. The result is a policy choice: to what extent, if any, should the old subsidize the young or the healthy subsidize the sick? Or, put another way, to what extent should sicker people pay premiums commensurate with their utilization of health care services? The ACA endeavors to create single, large state-wide pools, with strong cross-subsidies. The repeal and replace proposals instead favor a more fragmented market. That could work, but it would depend on how well states respond to the needs of higher-risk individuals through some form of reinsurance mechanisms or catastrophic risk pools aimed at protecting those populations.


The contentious debate over the fate of the ACA—whether to repeal or replace, or to “repair”—is nominally about coverage and premium costs. The ACA’s proponents seek comprehensive affordable coverage, but that has proven to be an elusive goal. The ACA’s detractors point to spotty coverage and rapidly rising premium costs as evidence of the ACA’s failure. They hope that by giving states wider latitude, market forces will furnish a solution that has to date evaded policymakers. In the meantime, although policy makers continue to express concerns regarding the underlying costs of medical care and the need to bend the cost curve, neither side has effectively addressed that issue.

By the end of the month, we will know whether the Republican effort repeal and replace the ACA has succeeded. Whether it does or does not will in all likelihood result in a good deal of partisan “crowing,” all of which we suspect will be unhelpful.

By: Alden Bianchi & Edward Lenz


Back to Top