A valuable proposition

ITIJ 198, July 2017
When underwriting an IPMI policy, key pricing and experience rating considerations must be made. Joanne Buckle and Neha Taneja share their experience
Group private medical insurance (PMI) protects employers and their employees against the costs of medical care. Traditional domestic PMI products operate in the country of residence, limiting the beneficiary to receive treatment in that country. However, with increased globalisation, many companies today need to provide uninterrupted health cover for employees with significant international long-term travel requirements. International PMI (IPMI) provides much broader coverage, with cover in locations outside of the beneficiary’s home country, giving seamless access to comprehensive healthcare services across the globe. 
IPMI may often be confused with travel insurance which is usually a short-term policy designed for holidaymakers or travellers on short business trips to cover cancellations, loss of personal belongings and emergency medical treatment. An IPMI policy is targeted at employees who have long-term travel outside of their home country. It typically provides cover for comprehensive medical treatment including medical evacuation and repatriation.
The global IPMI industry for expatriates and students was worth approximately US$13 billion in gross written premiums in 2015 and is expected to grow to $20 billion by 2019, according to Finaccord’s latest research. This translates to an annual growth rate of about 11 per cent. Taken from Pacific Prime’s Cost of International Health Insurance Report 2016, below is a list of the top 10 insurers that together total approximately 70 per cent of the total number of plans sold worldwide:
  • Aetna International.
  • AXA international.
  • IntegraGlobal.
  • ALC Global Health.
  • BUPA Global.
  • Now Health International.
  • Allianz Worldwide Care.
  • William Russell.
  • Cigna Worldwide.
  • InterGlobal (now Aetna International).
The market has become very competitive, containing many smaller and regional insurers, as well as large global insurers. As a result, profit margins, which have been generous traditionally, have come under increasing pressure. If domestic insurers want to expand into IPMI profitably, they must be able to price group policies appropriately, taking into account a number of increasingly challenging issues. 
This feature article looks at the current global market spread and size of the IPMI market and discusses some of the key considerations for pricing and experience rating a group IPMI policy.
Pricing considerations
Dealing with multiple geographies, changing regulations, various health systems, diverse demographics and movement of the insured population results in a number of additional complexities when compared to rating a traditional PMI policy. Here are some of the key factors IPMI providers need to consider:
Local data limitations: The wealth of data that a traditional health insurer holds on domestic PMI policies is usually insufficient for pricing an IPMI product, because:
o IPMI policies usually offer a more much comprehensive benefit package.
o Differences in the socio-economic profile of the target market, resulting in markedly different benefit features and claims experience. 
o Distinct claiming patterns due to the international nature of the benefits. 
o Variation in utilisation patterns by country and nationality. 
o Portability offered under an IPMI policy allows full access to benefits wherever the employees are and it is difficult to predict where different services will be consumed.
Obtaining reliable and relevant data with a desired level of granularity can be challenging making it difficult to get any credible results on which to base sound conclusions.
Geographical area of coverage: This is considered one of the key rating factors for an IPMI policy as claims costs can vary significantly between countries. For example, most insurers provide separate cover for ‘worldwide excluding US’ and ‘worldwide including US’, because healthcare costs are typically much more expensive in the United States than anywhere else in the world. Most insurers would classify countries into different regions/levels/zones that have broadly similar costs and healthcare systems for more accurate rating. However, constructing such classifications is difficult because: 
o Limited claims experience for some countries and lack of data for others makes the classification statistically less sound.
o Even countries with similar costs may have different types and quality of healthcare services, disease trends and state healthcare system,s which can make it difficult to group countries into particular zones. For example, insurers may experience lower claims ratio in countries with well-functioning state healthcare systems, which allow access for temporary residents. The rules on whether an overseas national is eligible to access the local state healthcare system are complex and vary by destination country, as well as nationality. In addition, the likelihood that an employee will access state coverage depends on the quality of the state healthcare system, as well as the nationality and cultural preferences of the employee. 
o Volatility in exchange rates can result in the pricing zone relativities becoming rapidly obsolete. 
All of these factors are likely to have a significant impact on the claim frequencies and costs. As a result, trying to price cover accurately for a multinational company with employees residing in multiple countries across the globe is quite a task.
One of the major problems we observe in IPMI business is the lack of emphasis on robust experience rating exercises
International medical inflation: While most insurers have established mechanisms to control domestic medical inflation, cost controls for IPMI are relatively difficult. In general, insurers have struggled to generate enough business volume for negotiated contracts or deep discounts. Therefore, the traditional methods for steering utilisation, such as networks, co-pays and utilisation management are less useful. With growing volumes, especially in the major hubs, provider network management is becoming critical to get the best prices. Increased use of case management and expert second opinion to review and offer a change of diagnosis or treatment is also helping minimise unnecessary treatments and, hence, costs. The cost containment strategies adopted by insurers should, however, give due consideration to employee satisfaction and take care not to compromise the quality of cover or service provided.
Regulatory reforms: Regulation is a key consideration and an area of increasing complexity for the entire market. The demand for medical treatment is growing and, with mounting pressure on the state systems, there are likely to be stringent healthcare laws in many jurisdictions. Insurers have to make sure their products are compliant with differing regulations in each country. Rules on underwriting and pre-existing conditions designed for domestic markets can have a severe impact on pricing for IPMI – often unintentionally. Some examples of recent regulatory changes that have had significant impact on the market are:
o Dubai Health Authority Reform has implemented mandatory health insurance coverage for all residents. While aimed at ensuring employees have sufficient domestic medical insurance so that they are not a burden on the state system, the regulations have had a significant impact on IPMI – resulting in some insurers withdrawing plans and many others having to adjust the coverage and premiums to comply with these reforms.
o China Insurance Regulatory Commission (CIRC) has issued more restrictions to curb the growing trend of mainland individuals purchasing coverage in Hong Kong by enforcing the prohibition of sale and marketing of Hong Kong insurance products in the mainland.
The changing local regulatory landscape often makes it hard for an IPMI insurer to build a robust and scalable pricing model that will work in multiple jurisdictions, which leads to increasing administration expenses.   
Economic uncertainty: Challenging economic conditions and significant swings in exchange rates result in the historical data being less relevant to future experience. For example, the falling value of sterling post-Brexit has changed relative global claims costs and hence profits for some IPMI insurers.
Target market: The target market is continuously evolving with focus gradually shifting from the West to the business hubs in Asia and the Middle East. The market is also seeing a growing demand from high net worth individuals for comprehensive health insurance plans with modular benefit designs. Millennials are also likely to be a viable target market due to their desire to live and work abroad. These shifts in target markets could impact future product designs and hence pricing of these policies. It is, therefore, important to keep track of emerging markets in order to maintain desired sales and profitability.
The target market is continuously evolving with focus gradually shifting from the West to the business hubs in Asia and the Middle East
Innovative product design: Changes in the target market have led to different expectations from the IPMI product in terms of having access and coverage. The aforementioned regulatory changes and increasing competition require insurers to work continuously towards the development of more innovative and improved products that are also locally compliant. Developing a hybrid product between domestic and international healthcare policies designed to suit target locations might be the answer. Offering a modular design allows the companies and the brokers to cherry pick the benefits to suit their needs and budget. This may, however, increase the risk of anti-selection. Also, there has been an increased focus on value added propositions and wellness benefits but it is often difficult to scale and provide these globally. 
Nationalities: The target market for this product tends to include a cohort of people with different nationalities residing in different parts of the world. Nationalities have often been observed to have an impact on the propensity of claiming, which could significantly affect the claims experience if not accurately reflected in the rating exercise. However, it is often difficult to get credible numbers at this level of detail. 
Experience rating
As business volumes are growing, insurers are building up claims and exposure data for their IPMI business. For larger groups, the insurer can then use the group’s claims experience to estimate a more accurate price by experience rating. Experience rating is the process by which a group’s historical claims experience is reflected in the future rating calculations. This can be done prospectively, by projecting the future claim costs using historical claim experience, or retrospectively, by providing a ‘profit’ share to the employer based on the claims experience during that period. 
The experience rating exercise assumes that the past experience gives a reasonable expectation of the future, while making allowance for factors such as statistical fluctuations, trends, changes in exposure mix, turnover and benefit changes that could result in future claim costs that vary from the historical experience. 
Some of the issues that we have observed during IPMI experience rating exercises are shown in Figure 1.
One of the major problems we observe in IPMI business is the lack of emphasis on robust experience rating exercises. Often group size is not given due consideration when assigning credibility to the group’s experience. Basing the rates on data with higher statistical variation could result in significant under or over-estimation of the premiums with consequent loss of business or loss of margin. Also, margins for large groups in particular are being cut and products largely commoditised in a highly competitive market. This leads to behaviour such as inconsistent use of large claims pooling charges, removal of years of poor experience or large claims as ‘anomalies’ and inconsistent use of trend and benefit factors. 
Successful application of group experience rating requires considerable actuarial and underwriting judgement, combined with a robust model and a strong, auditable process with high levels of discipline and governance.
Constant evolution
We are now entering a phase where IPMI has gradually moved from being a ‘luxury product’ to a ‘must have’ as international lifestyles continue to expand and evolve. However, cross border healthcare is a complex issue. Insurers therefore need to have the global expertise to provide plans that function well in multiple markets yet provide the flexibility to allow for localised customs and practices.
To be able to develop and price a product that covers globally and delivers locally, the insurer, while giving due consideration to client desires, needs to keep a close watch on global and regional developments in the market. It needs to continually evolve the benefit offerings to meet changing business needs and landscapes. As world healthcare standards continue to rise and the range of treatment options available continues to increase, unit costs and utilisation will continue to increase. The insurer can limit the impact of rising costs through robust streamlined cross-border claims systems, global network management negotiations and strict anti-fraud measures, thus offering a valuable proposition to employers. 
Sustainable premiums and cost containment are critical concerns for the longevity of this market. Achieving a balance between pricing and suitability of coverage and benefits, keeping in mind regulatory requirements, is the core challenge at the moment. Insurers will have to get pricing right and look to distinguish themselves from the competition through innovation and the inclusion of value-added benefits and services to stand out from the crowd. For those that can build a robust and scalable service model offering high levels of value to the customer, the potential margins are significant. ■
Figure 1
Issues Definition Problem Area
Credibility Credibility thresholds assigned to a particular group’s experience for the calculation of final rates are usually based on the number of members covered under the scheme. The bigger the group, the higher the credibility assigned to the group’s experience. This idea is based on a principle that the statistical variation in expected claims cost tends to decrease as the group size increases. In the event of non-availability of historical group data, or if data is considered to be below the minimum threshold, manual rates are used. The product features and the international nature of the policy introduces a lot of heterogeneity in membership profiles and claims experience. The data is, therefore, more susceptible to higher statistical variation. Hence, we expect the credibility thresholds for IPMI business to be much higher than domestic business, but this is not always reflected in local business practice.
Large claims pooling Large claims pooling is a system designed to help stabilise premium fluctuations in smaller groups. Large claims over a stated amount are charged to a pool contributed to by many small groups that belong to and share in that pool. The smaller the group, the lower the thresholds for pooling level. Larger groups will have much higher thresholds for large claim pooling. Large claims could result in excessive rate increases driving away good business, impacting business volumes and overall profitability. Isolating the large claims and charging them separately to the pool not only helps defray the large claims, but also protect the groups against catastrophic rate increases and insurers against high lapse rates. However, for commercial reasons, not all insurers adhere to the discipline of pooling the large claims over the entire portfolio. 
Data There is always a trade-off between the volume and relevance of data that can be used for experience rating. Data that is too old is likely to be less relevant for projecting future experience. It has been observed that the predictive value of experience data decreases over time due to changes in trends, reduced correlation of the costs between one period and the next and changing demographics. The problem with using very recent data is that it may not be fully developed, which introduces further uncertainty to the estimates. A compromise therefore needs to be made between using old and recent data to make sure that the dataset used is not only credible but also relevant and reasonably developed to minimise volatility when developing the rates.
Trends It is necessary to allow for appropriate trends in the claim frequencies and costs to more accurately project the future experience of the group. Heterogeneity in the underlying data may result in undesirable random variation being captured in the trends developed from the past claim experience. It is therefore, important to validate these trends objectively with a reasonable view of the future.