23 June 2014Actuarial & underwriting

Innovation in reserving methodology

Healthcare providers were among the earliest adopters of the captive model for self-insurance. The medical malpractice crisis of the 1980s (revisited in the early 2000s) prompted many healthcare systems to take significant self-insured positions that have grown over time. Bermuda was among the first domiciles, in the early 1980s, to encourage such captives, and these healthcare industry medical malpractice captives have performed well. Despite an extended soft market, medical malpractice captive formations continue to this day.

Recent changes to the US healthcare payment and delivery system have put pressure on healthcare providers to adapt to declining and changing reimbursement models. Cost reduction, innovation and adaptation are fundamental to the long-term ability of a healthcare system to ride the wave of change rather than get swept under by the current.

The impact on medical malpractice liability of the Patient Protection and Affordable Care Act (PPACA) is difficult to quantify. Many expect an increase in claims frequency as millions of new insureds re-enter the healthcare delivery system. A RAND Corporation study, How Will the PPACA Affect Liability Insurance Costs? published in April 2014, estimates a nationwide average medical malpractice increase of 3.4 percent.

Beyond liability costs, the risks of forming and operating an Accountable Care Organization (ACO) or an ACO-like organisation are numerous and evolving. Some of these risks are traditional malpractice, such as clinical integration; others lean more to vicarious liability, such as care coordination and health and wellness coaching.

A risk finance programme with gaps in coverage adds hidden costs to the parent and diminishes its ability to compete successfully. Conversely, a bloated programme with reserves and capital that far outstrip potential insured liabilities carries a heavy opportunity cost to the parent organisation. Captive operators should focus on matching the insurance and risk services provided to the parent to the overall financial strength of the captive.

No longer can the success of a healthcare captive be measured by another year of declining excess premiums; healthcare parent organisations need more. A risk finance programme that effectively responds to these challenges will provide its parent organisation with a competitive advantage.


The core function of the captive insurer is to pay claims. To that end, captives and their service providers face the challenge of setting accurate case reserves. Most often, the responsibility for determining the ultimate value of a claim or lawsuit lies with the insurance claim adjusters. These adjusters look at the facts of the case in light of past experience, the opinion of their experts, and potentially what other similar cases have resolved for, and give their best educated guess of what the eventual value of the case will be.

While similar fact scenarios may turn up from time to time, the truth is that no two cases are alike. With multiple factors driving exposure, the complex information that needs to be weighed, and the added uncertainty of what a jury may decide to award, intuiting the true ‘value’ of a claim is enormously difficult.

Too little held in reserve can result in an embarrassing situation for the risk manager, general counsel and CFO, and to their explaining to the board why a certain case was reserved at a fraction of the ultimate claim value. Under-reserving threatens the good financial planning and operational excellence pursued by the organisation 365 days a year.

The dangers of over-reserving are more subtle. Risk management sets individual case reserves and is motivated by a ‘no surprises’ mantra from management. The motivation is self-evident—a claim that is over-reserved results in a happy financial variation when its ultimate value is less than the reserve. No embarrassing conversations with management or the board ensue and everyone is happy. But should they be?

There is a hidden opportunity cost to over-reserving that gets multiplied by conservatism built into every step of the self-insurance process. A (conservative) case reserve is set by the risk manager, third party administrator and general counsel, the actuary provides a (conservative) report based on the reserved amounts, and the captive board selects a (probably conservative) confidence level and discount rate.

Over-stated reserves put pressure on capital, increase loss funding estimates for subsequent years and encourage reinsurers to charge higher premiums for excess coverage. In the end, higher booked reserves necessitate more capital resulting in potentially millions of dollars locked up that could be better spent pursuing the strategic objectives of the organisation.

The 2008 recession encouraged even more conservatism by pushing down or eliminating the use of discount rates. While this may be an accurate decision in a vacuum (after all, what rates of return are captives really earning in today’s interest rate environment?), on the whole the reduction of discount rates added to the overall conservatism of the reserving process.

A novel approach

TRA identified reserve setting as the linchpin for an effective, efficient self-insurance vehicle and went hunting for an evidence-based, robust, and transparent methodology that would enable stakeholders to understand the information, risks and the values that went into the reserve setting decision-making process.

TRA developed a novel approach known as Decision Analysis Reserve Targeting (DART), which brings to bear the tools of decision analysis and the enterprise risk management (ERM) framework to the reserve decision-making process.

For those unfamiliar with the field, decision analysis is an academic discipline pioneered at Stanford and Harvard Universities in the 1960s and 1970s that is concerned with addressing the challenges of making high-quality decisions under uncertain situations. The ERM process provides a scaffold for ensuring robustness and an iterative process for continual improvement.

Rather than take the process out of the hands of the expert claims adjusters, DART leverages their expertise and addresses potential errors by guiding them through a scientific and structured decision-making process of:

1.  Developing a model of all of the decisions, risks and uncertainties in each claim or law suit;

2.  Applying high-quality Bayesian subjective probability assessments to all the possible outcomes; and

3.  Calculating the mean liability, known as the expected value, which then becomes the case reserve.

By dividing the decision-making process into simple pieces, the DART process provides:

•   Clear guidance to the organisation for setting the loss reserve on each case, leveraging the expertise of decision-makers, while removing the guesswork of the standard process;

•   Accurate, early assessment of the liability and exposure in each case;

•   Optimisation of the amount held in reserves versus payouts, addressing the issues of over and under-reserving;

•   High confidence in the aggregate held in reserve and, therefore, the robustness of the insurance programme;

•   A transparent, thorough, and defensible understanding of the issues and exposures in a case; and

•   Unique depth of insight that provides clarity for communicating and driving case strategy.

TRA has been using DART for Stanford University Medical Network medical malpractice claims for three years and the early results are promising in expected, and unexpected, ways. Perhaps most important—though difficult to quantify—is that working through the DART process with the claims team informs case strategy.

A case study

Consider a sample case using traditional case reserving and then using DART.

The claims team considers a case ‘very defensible’ and expects to win at trial; the plaintiff’s lawyer offers to settle the claim for $150,000.

Given this information, most self-insurers would decline the offer to settle and continue to defend the case.

Using the DART process new information is rendered. After careful calibration and discussion led by a DART expert, the claims team sets the likelihood of a plaintiff verdict at 25 percent. If the jury does return a verdict in favour of the plaintiff, it is expected to be valued at $1 million.

Understanding this information, the DART process establishes an expected value of the case of around $250,000. Now, accepting the settlement appears prudent and would save the parent $100,000 even though a defence verdict is likely at trial.

The early quantifiable results of utilising DART are promising. Loss development factors over five years for medical malpractice claims have been reduced significantly. Before the implementation of the DART process a $1,000 claim would develop to just over $4,000; post DART that same claim now develops to just over $2,500 (see Figure 1). This means that once a reserve is set on a case it needs to be adjusted much less than the industry average.

The DART process engenders confidence in reserves for all participants in the self-insurance system: risk managers, general counsel, CFO, actuary and the captive board. This confidence reduces the need for additional risk margin for adverse loss development. Ultimately, confidence in reserves and sound understanding of capital needed to support them free the captive to respond to the new and emerging risks faced by the parent as the PPACA rollout continues.