Risk Pooling: What is fair?
Authored by Nick Reilly, Head of Business Development, UK and North Europe, RNA Analytics
There has been considerable discussion on what is considered ‘fair’ when we are talking about insurance:
Example 1: Should car insurance premiums be either so expensive, or unavailable, in some areas, preventing individuals the ability to have their own car?
Example 2: Should protection insurance (life, CI or IP) be unavailable for those individuals who have certain medical histories?
Example 3: Should some houses be uninsurable, due to flood risk, meaning the individuals who own them are unable to protect their own property?
In essence, what is ‘fair’ is determined by society, although there is an indirect route via our elected officials or regulators. The issue of flooding (example 3 above) was suitably high profile to lead to the establishment of Flood Re in 2013. In simple terms, all UK households, via their insurers, fund a pool to pay for claims in high risk areas. Could this be expanded to other areas?
The Protection example.
Could this work for the example above for protection insurance (life, CI and IP)? This isn’t as clear cut. While all individuals who own a home in a flood area require cover for flood damage, not all individuals need protection insurance. Therefore, those who would have been declined (insurers refusing to cover them) due to their medical history are not necessarily disadvantaged. Examples of this include:
The many (but not all) working individuals who get some protection benefits as part of their employment package. This partially, or in some cases fully, meets this need while employed.
Individuals who have no dependents or need for a lump sum or income (such as many retired individuals) do not have a need for any cover.
Those individuals who already receive benefits from the state, given their underlying health conditions (although this may be at a subsistence level).
In the UK, for example, it is hard to see any government prioritise a form of ‘Life Re’ given the economic climate, alongside funding pressures for the NHS and benefit payments.
As most economies have a robust and competitive market for protection insurance, cover is available for many prospective customers. However, if there is no ‘Life Re’ safety net, then individuals who are ‘declined’ run the risk of economic hardship if the worst happens. There are some ways that this could be mitigated (to some extent), such as:
Buying policies earlier on, when the customers are younger (and so in better health), although this requires early financial planning, and possibly increased costs by paying for the plans for longer.
Taking specialist advice to find companies that offer tailored plans, or policies that are designed for customers with certain medical conditions.
Accepting plans with exclusions on certain causes, which offers cover on the claim causes that were insurable.
None of these are ideal, but offer a possible route to reduce the economic loss from death or disability.
For the past generation, the process of risk acceptance has become more complicated, with more underwriting rigour. This has led to a reduction in the homogenous risks within each pool (in other words, individuals are considered different levels of risk depending on the underwriting information). This trend was due to two primary reasons:
To better understand, and price for the risk, so that the insurers opened themselves up to less risk.
As part of a drive to reduce the headline market price for the ‘best’ lives, to order to compete in a price driven industry.
Regulation and a wider risk management regime have driven the first reason, yet the second is due to protection as being seen as a price sensitive commodity.
Unfortunately, once one player makes a price reduction by increasing underwriting rigour (so moving more customers out of ‘standard’ pools, or increasing ‘decline’ rates), others must follow to prevent being selected against.
This competitive market will continue to reduce the risk pool. Some insure-tech entrants use this as a goal, and suggest that their ideal would be to have a price based on the exact risk criteria of each individual. Unfortunately, the logical end state to this situation is the demise of risk pooling. By this point, cover is incredibly cheap for those with the lowest known risk factors, and unavailable for those who would previously have been cross subsidised.
To prevent this, restrictions could be put on the ability of insurers to continue to add ever more risk factors or use data to split the remaining risk pools. This is not an easy task, but there are possible approaches that could be pursued to put this in place.
Our decision makers on behalf of society (our elected officials, and in some degree, our regulators) should be aware of this and either take action before it is too late or accept the consequences.