Challenges of pricing health insurance products in a Covid-19 environment


A question we have received quite a bit lately is: How do we price health insurance products now, knowing that Covid19 is around? In some cases the question related to incorporating Covid19 coverage into the product, in others it was just to understand how claims costs will change due to Covid19. In a couple of requests it was to directly price a Covid19 product. This article will focus on this question. First some perspective; when we discuss Covid19 and insurance we could be talking about a wide range of coverage, including:

  • The effect of Covid19 on death coverage, as mortality is affected for older ages especially as well as substandard risks and whether this is simply short term volatility or is this a new reality. 
  • The effect of Covid19 on general insurance coverage such as business interruption, motor coverage (people stuck at home don’t get into car accidents) and other products.
  • The effect on health insurance, whether this is hospitalization coverage, major medical, critical illness coverage or coverage directly for Covid19 (hospitalization, temporary loss of work, critical illness and / or death).

The challenge we face for pricing health insurance in a Covid19 environment is twofold:

  • The high level of volatility in incidence rates of Covid19 and the effect of government actions and level of discipline of the population.
  • The interaction effect between Covid19 and other types of coverage.

The volatility in incidence rates of Covid19 is well known and documented. For some time Malaysia was in the top ten in terms of number of Covid19 cases but now with the movement control order in place for over a month there are now 50 countries with a higher number of cases. There is quite a bit of statistics available online. For instance the website worldometers.info tracks fields such as:

  • Total as well as new Cases and number of deaths
  • Total cases as well as deaths per 1 million population
  • Total tests as well as tests per 1 million population

There are also graphs showing the growth in cases over time. Take the example of Malaysia versus Singapore. Malaysia has a significantly higher population, yet the number of cases in Singapore is triple that of Malaysia. However in both Singapore as well as Malaysia there are segments of the population which are at a much higher risk, such as the migrant workers in Singapore. In Malaysia as well there are states which are classified as ‘green’ and thus low risk whereas in Kuala Lumpur the classification is consistently ‘red’.  Thus if we are pricing a health insurance product in Singapore, it is very unlikely that the market segment being targeted includes migrant workers. In Malaysia however, Kuala Lumpur is a major target market for insurers. Beyond just segmenting the population there is also the issue of the extent of testing being done. Some people have Covid19 but they do not show symptoms, so they potentially pass on the virus to others. In Singapore 24,600 people are tested per 1 million population whereas in Malaysia it is only 6,050 per 1 million population. Will greater testing result in greater numbers of Covid19 cases in Malaysia?  Linking this back to insurance: if Covid19 will be eliminated by say June this year in a country then the effect of Covid19 is likely to be short term indeed, but if the effects linger on due to continued breakouts then there will be a larger cost involved.

Going beyond Malaysia and Singapore, there is a perception that in Indonesia relatively little is being done in terms of government intervention. Thus although Indonesia officially has just over 11,000 cases, testing is only 413 per 1 million population. It might be reasonable to expect that the situation is likely to get significantly worse before it gets better (and is much worse currently than the numbers indicate). In Bangladesh the government has taken action such as restricted movement, but the perception from the news is that the population has partly ignored such requirements, with an example being the death of a well-known religious leader whose funeral had a huge gathering of people. Bangladesh has almost 9,500 cases but again testing is low: 494 per 1 million population.

The second challenge with pricing health insurance in a Covid19 environment is the interaction affect with other coverage. For instance with a restricted movement order it is less likely that people will get into accidents such as on the way to work, and even if there are medical problems, it is likely that policyholders will try their level best to avoid being hospitalized due to fear of being in proximity to people who have Covid19. With respect to critical illness coverage there is the fear (and emerging data as shown in a recent article in the Washington Post) that people who have had Covid19 are more susceptible to critical illness such as stroke. What was particularly worrying is that the surge in patients involved were relatively younger and in good health. They were barely sick with Covid19, but yet dying due to strokes. Also worrying is that whereas major medical and other such coverage tends to be yearly renewable, meaning the premiums could be increased next year if conditions warrant it, for critical illness coverage the pricing might be with explicitly or implicitly guaranteed premium rates.  For any product which will include Covid19 benefits we will need to determine the extent of anti-selection. For instance, it is logical that front liners such as doctors, nurses, food and other delivery will strongly desire such coverage, but in this case is this considered anti-selection? Their exposure to risk is definitely high but they are also the most aware of proper hygiene and other practices to minimize risk.  Thus does it make sense to exclude these risks or other risks such as frequent travelers? Waiting period will also be an issue, as policyholders will want immediate coverage but it does take a bit of time for the symptoms to appear. When designing benefits we need to keep the big picture in mind: the goal to reduce risks through proper risk management. Thus we would prefer to focus payouts on those policyholders who for instance must enter ICU as this is where the higher costs are incurred, as opposed to those policyholders with no or slight symptoms.

It is possible (and necessary) to continue to price insurance products in this Covid19 environment. It is up to us to wade through the data to determine how much is temporary volatility and how much is the new normal.  We would be happy to discuss this in more detail with you, at Hassan.Odierno@Actuarialpartners.com or https://www.linkedin.com/in/hassan-scott-odierno/

Practice Areas: