Genetic Testing and Insurance:
The Complexity of Adverse Selection
Maureen Durnin, Guelph, Canada
Michael Hoy, University of Guelph, Canada
Michael Ruse, Florida State University, USA
ABSTRACT. The debate on whether insurance companies should be allowed to
use results of individuals’ genetic tests for underwriting purposes has been both
lively and increasingly relevant over the past two decades. Yet there appears to
be no widely agreed upon resolution regarding appropriate and effective regulation. There exists today a gamut of recommendations and actual practices
addressing this phenomenon ranging from laissez-faire to voluntary industry
moratoria to strict legal prohibition. One obvious reason for such a variance in
views and approaches is that there are competing bases for evaluating the outcomes of restricting insurers’ use of such information. For example, an outright
ban on the use of genetic test results may seem the best method for protecting
against unfair discrimination, while allowing their use may seem to be the best
way to foster efficiency in the market for insurance. However, there is also a
lack of understanding about how restricting the use of genetic information
would play out in the market through the so-called phenomenon of adverse
selection. Using economic analysis, we discuss how the type of adverse selection
that occurs in insurance markets affects various arguments both in favour and
against an outright ban on insurers’ use of genetic tests for pricing insurance.
We review arguments based on moral principles (i.e. a concern with unfair discrimination as well as welfarist analysis related to distributive justice). Practical
concerns from the insurance industry based on actuarial principles and economic efficiency are also compared. Each perspective is shown to lead to a
range of conflicting recommendations about how genetic information should be
regulated and these conclusions depend critically on whether one conducts the
analysis from the ex ante temporal perspective (i.e. before individuals know their
risk type), from the interim temporal perspective (i.e. after individuals know
their risk type but before they purchase their insurance policies), or from the ex
post temporal perspective (i.e. after all uncertainty is resolved, including the
claims status of each policyholder).
KEYWORDS. Insurance, genetic discrimination, regulation
ETHICAL PERSPECTIVES 19, no. 1(2012): 123-154.
© 2012 by Centre for Ethics, KULeuven. All rights reserved.
95288_EthPersp_2012/1_07_Durnin.indd 123
doi: 10.2143/EP.19.1.2152682
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
I. INTRODUCTION
F
rom the beginning of the Human Genome Project in 1990, and perhaps even prior to that time, there has been much discussion and
disagreement about how to use the emerging information provided by the
increasingly detailed mapping and identification of specific genetic markers in human DNA. This progress in genetic testing, while fascinating and
productive for the scientific community, produces potentially chaotic
effects on insurance markets. Global response to genetic testing has
resulted in the adoption of many different regulatory frameworks which
vary significantly between countries and sometimes within countries.
Many OECD countries have similar protections for personal privacy and
against religious, racial, and gender discrimination. Thus, although there
may be justifications for variations, one may reasonably expect such
countries to have similar protections against genetic discrimination. In
addition, over time, the quantitative importance of genetic testing to
underwriting and pricing of insurance is likely to become increasingly
relevant. The questions arise: Why is there continuing debate? Why do
countries that otherwise adopt similar social systems and similar protections for personal privacy and against racial, religious, and gender
discrimination choose different legislation regulating access to genetic
information and potential genetic discrimination?
The academic literature addressing the problem offers many conflicting viewpoints. The difficulty arises from the complexity of the multiple
reactions at play in the consideration of the ultimate use of genetic testing. There exists a wide variety of foundations from which stances are
based, ranging from purely moral principles to criteria based more on
pragmatic concerns about how markets satisfy the needs of individuals
and firms to manage risks. Ethical arguments are typically concerned with
issues of potential discrimination and loss of privacy rights. The business/
actuarial based arguments address the functionality of markets for various
types of insurance. The welfare economics approach to the argument
— 124 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 124
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
includes considerations of efficiency and equity over resources. An important factor that may complicate the arguments based on moral principles
as well as other perspectives based on more pragmatic considerations is
the variety of market outcomes that are possible when individuals hold
increasing amounts of private information. When bans stop insurance
companies from using genetic test results to evaluate risk and set prices
accordingly, individuals are free to obtain and privately hold genetic information that may be pertinent to their insurance purchasing. Economists
refer to this phenomenon as adverse selection, the effects of which will
become evident later in the discussion of pooling versus separating equilibria.
Differing temporal perspectives also exist and have to be considered.
Should a moral principle or other criterion be applied from the ex ante
perspective (i.e. taken at the point in time before people know their
genetic types) or from what we will call an interim perspective (i.e. taken
at the point in time after people know their genetic types and make their
insurance purchasing decision). One can also assess the appropriateness
of a regulatory ban from the ex post perspective (i.e. once all uncertainty
has been resolved and insurance claims have been processed).1
In the present contribution, we propose three very different sets of
considerations that interact in a complex manner when assessing the efficacy of a ban on the use of genetic test results by insurers: (i) one must
decide on which ethical or other principle(s) to apply to the policy question, (ii) one must determine what the outcomes are likely to be of such
a ban in terms of market outcomes, and (iii) one must choose a particular
temporal perspective to adopt in making relevant assessments of outcomes. We highlight the relationships among these three considerations
by making a comparison between a laissez-faire approach (i.e. allowing
insurers to use whatever information their clients have obtained from
genetic tests) and a regulation that bans insurers from using any such
information. Such a stark difference in policy options brings out clearly
the complexities involved in the academic and public debates as well as
— 125 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 125
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
the practical difficulties in resolving this issue. We believe that these
factors, when considered in the context of how people perceive the implications of genetic information having an impact on their sense of privacy
and overall well-being, explain why it has not been easy to resolve this
important public policy issue. The attractiveness of different ethical and
other criteria turn out to be context-specific and so it is not surprising
that a clear and widely acceptable legal framework has appeared elusive.
Our goal here is to explain the complexities involved in the various arguments. We do not propose to offer a definitive solution or even to weighin significantly on the relative strengths of the various moral and other
foundations per se. We do, however, demonstrate that in certain economic
environments, as defined by the type of insurance contracts and the quantitative importance of the genetic information that people hold, there are
some arguments that have little to no relevance or force. In this way, we
also provide some guidance for future discussion about the choice and
design of an appropriate framework for resolving this important question.
II. WORLD VIEW
Across the globe, the reaction to the use of genetic testing for life, private
health, and long term disability insurance purposes varies from legislation
or total moratoria, which bans any use of genetic test results by insurers,
to a status quo approach, letting the industry regulate itself. In most of
Western Europe the ban is almost total, falling in line with the UNESCO
Declaration on Human Genetic Data 2003. In Belgium, insurers are prohibited from accepting even favourable genetic test results provided
voluntarily by consumers. In the United Kingdom and the Netherlands,
companies can ask for genetic test results only for large policies (those
exceeding £500,000 in Britain and the approximate equivalent of $150,000
US in the Netherlands. The latter is adjusted every three years to the cost
of living.) In Britain, the types of genetic tests that insurers can request
— 126 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 126
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
for policies exceeding the cap are restricted to those deemed relevant by
an independent committee. In Asia and the Middle East, with the exception of South Korea and Israel, who both ban genetic testing for insurance purposes, no specific stances have been taken on genetic testing use
for insurance purposes. In South America and Africa, only Brazil and
South Africa have begun to address the issue. Brazil requires the insured’s
consent and South Africa retains a status quo approach. Australia, New
Zealand and Canada are also among those who allow the status quo to
remain, relying on existing privacy laws and the insurance industry’s assurances of fair self-regulation. The United States is a particular case in that
the discussion there, in the absence of socialized medical insurance,
involves both the health and the life insurance industries. Regulations
likewise vary from state to state. Federally, the Genetic Information Nondiscrimination Act (GINA) – passed in May 2008 – addresses the use of
genetic testing in employment and health insurance, though only 14 states
have introduced some laws to govern the use of genetic testing in life
insurance and these laws make restrictions rather than outright bans.2
III. COMPETING MORAL PRINCIPLES AND OTHER PERSPECTIVES
So why are so many countries opting for strict regulations (and often
total restrictions) on the use of genetic tests to evaluate risk levels and
formulate prices in the insurance industry? Legislative bodies are
responding to public opinion and public opinion is strongly against
insurance companies requiring genetic test results in order to separate
people into risk categories.3 These reactions can, in part, be attributed to the
moral principles governing privacy. The genetic code is perceived as the
very essence of the self. As Paul Root Wolpe points out in his comments
on Dorothy Nelkin and Susan Lindee’s The DNA Mystique: “We have
biologized the idea of the soul, the essence of each person, that which
can recreate us and which encapsulates all that we are.” He continues:
— 127 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 127
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
“Genes are, in some sense, more truly us than we ourselves are; they are
us condensed, boiled down, reduced to essence” (1997, 216-218). Even
monozygotic twins have only nearly identical DNA as differing environmental influences in the womb and throughout their lives affect which
genes are turned on and off. This process is called epigenetic modification. As a result, an individual’s genetic code is arguably the most personal of possessions. Being required to reveal one’s DNA through
genetic tests, strips away any possibility of retaining sole ownership of
one of the most basic aspects of one’s being. Direct-to-consumer companies offering DNA testing underline the idea of providing genetic
profiles that reveal information that will benefit consumers by not only
offering predictions for health but also “fundamental insights into their
identities” (Nordgren and Juengst 2009), a concept referred to as ‘essentialism’. Whether the essentialist view is an appropriate one or not is not
our focus here. It is merely important in pointing out that the general
public has, to a significant extent, adopted this view and legislators
around the world are responding to it.
The desire that others not be able to discover an individual’s personal
information, an intrinsic sense of privacy, is important to many. Combined with this is the fear that the results of genetic testing may cause
emotional and/or financial stress. What if the results mean bad news for
the individual and his or her loved ones? What if the outcome is that the
boss learns of unfavourable test results and denies promotion or even
terminates employment (and in the case of the United States, before the
so-called Genetic Information and Nondiscrimination Act signed into law
in 2008, access to comprehensive healthcare)? What if the results mean
insurance rates go sky high or, worse still, insurance is denied altogether?
This is the argument put forward by many support associations, such as
the Huntington Society of Canada, which were formed to raise the public awareness of specific diseases in the hope of raising funds for research
into cures and offering sufferers and their families counselling, aid and
general moral support.4
— 128 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 128
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
What if individuals can use genetic information about predisposition
genes to alter their lifestyle or medical choices? Any decision about regulating the use of genetic test results by insurers can have important spill
over effects beyond insurance decisions. For example, allowing insurers
access to genetic test results may create a disincentive for individuals to
obtain such information when it would be valuable to them for the purposes of deciding on important healthcare decisions (e.g. such as increased
frequency of colonoscopies should a person discover he or she has a
genetic predisposition to colon cancer).5 Not only is information of this
sort advantageous from the individual’s perspective, but it may also lead
to reduced costs for healthcare providers, be they public or private
(Fillipova and Hoy 2009). People may find such personal genetic information valuable, even if it would have no impact on health strategies since
knowing one’s genetic predisposition to diseases may alter optimal
savings decisions or affect procreation decisions.
There is also the case where, for emotional reasons, the individual
has no desire to learn of possible negative outcomes of genetic tests.
Research exists that underlines the individual’s right not to know the results
of genetic tests. “There is substantial evidence that troubling psychological effects can result from obtaining such information. For example, in
various clinical testing environments, when anonymous genetic tests were
offered at zero cost, Meiser and Dunn (2000) found that the percentage
at risk who requested testing varied from 9% to 20% in various centers
in UK cities and Vancouver. Obtaining information about a serious future
illness that cannot be avoided has been compared to learning about a
death sentence. People may also often simply prefer not to know information which may negatively impact on their life choices. Although it
may seem rational to want to have information about how an important
life activity will be affected in the future, people may prefer to leave their
future more open-ended” (Lemmens et al. 2008).6
Underlying these reactions about insurers and other third parties’
use of personal genetic information are the questions of need and right.
— 129 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 129
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
Do people need health and life insurance? Do people have a human
right to health and life insurance? Article 25 of the Universal Declaration of Human Rights drafted by the United Nations Commission on
Human Rights and adopted in 1948 declares: “Everyone has the right
to a standard of living adequate for the health and well-being of himself
and of his family, including food, clothing, housing and medical care
and necessary social services, and the right to security in the event of
unemployment, sickness, disability, widowhood, old age or other lack
of livelihood in circumstances beyond his control.” This declaration
clearly supports the belief that humans have a right to both health and
life insurance.
Let us consider the two kinds of insurance separately, beginning with
health insurance. Norman Daniels and Dan Brock, working with former
American President Clinton’s Health Care Task Force, composed a list
of principles and values as a basis for healthcare reform in the United
States. The first of these principles is: “Healthcare is of fundamental
importance because it protects the opportunities open to us to pursue
goals in life, reduces our pain and suffering, prevents premature loss of
life, and gives us information we need to plan our lives” (Daniels 1994,
426). The second principle on the list maintains that healthcare should
be universally accessible. While calling into question how Daniels arrived
at a principle of universally accessible healthcare, Shlomi Segall concurs
that healthcare is necessary and “is justified by way of society’s obligation
to reverse individuals’ disadvantaging biological conditions for which they
are not responsible” (2010, 171). It is important to underline here that
we are discussing healthcare and not health. The right to health and the
right to healthcare are two distinctly different questions. As Sen points
out: “Health equity cannot be understood in terms of health care [italics
original]” (2002, 660). In surveys performed by Rothstein and Hornung
(2004), consumers were asked to respond to statements of ‘right’ and
‘need’. To the statement ‘everyone needs health insurance’, the response
was 91.2% in agreement. When the statement was ‘everyone has a right
— 130 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 130
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
to health insurance’, the response was 90.6% in agreement. In our consideration of healthcare insurance, while pointing out varied philosophical
thought, it remains very appropriate to take the public’s opinion into
consideration since it is ultimately public opinion that drives decisions
made by elected lawmakers. It is also important to add that in most
OECD countries access to health insurance is essentially a moot point
since universal healthcare is already a fact.
When considering life insurance, again the 25th article of the Declaration of Human Rights is pertinent (though it is telling of the time that
that the men drafting the article refer only to ‘widowhood’). Death leads
directly to a loss of livelihood that is ‘beyond control’ of the individual
and though it may well be a female spouse’s demise that leaves that family without ‘security’, the result is the same. Martin O’Neill describes
life insurance as a social good “rather than being a pleasant luxury item.”
He continues by asserting that “Life insurance facilitates certain forms of
economic and social activity, such as the kind of entrepreneurial activity
in which individuals can engage if they have access to large and reasonably-priced bank loans; it allows access to the housing market; it facilitates
stable family life; and it provides for long term planning…” He also refers
to Article 16 of the Universal Declaration of Human Rights, which
declares the family a “fundamental group unit of society […] entitled to
protection by society and the State.” O’Neill goes on to suggest the need
for a “universal, solidaristic provision of access to life insurance” (2006,
578). His assertion that there is a (human) right to life insurance is definitely in line with results of the previously noted survey by Rothstein and
Hornung (2004). When the statements suggested to the survey participants concerned life insurance, 69.2% of respondents were of the opinion
that life insurance is a need.7 82.6% of those responding agreed that
access to life insurance is a right. If genetic test results lead to interference
with the public’s perception of what constitutes needs and rights with
regard to insurance, then it is important to create regulations that will
afford the appropriate protections.
— 131 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 131
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
In addition to human rights based arguments concerning privacy and
equal access to health and life insurance that favour a ban, there is also a
belief that people should not be punished in the market place for reasons
over which they have no control. This view, which contends that charging
different prices for insurance policies on the basis of genetic inheritance
is unfairly discriminatory, is also commonly held (e.g. Alper and Natowicz
1993; Kitcher 1996) and the expression ‘genetic discrimination’ has
become widespread. This perspective sits well within the general philosophy of Luck Egalitarianism. Although there are many nuances within that
literature, we focus here on Dworkin’s work.
In arriving at moral arguments about how economic relationships
should be governed, luck egalitarians distinguish between the roles of
brute luck and option luck. They argue that it is unjust to allow brute
luck, which arises from circumstances outside of an individual’s control,
to lead to differential outcomes (e.g. such as differences in the price of
insurance or more generally in standards of living). On the other hand,
luck egalitarians argue that people are responsible to make prudent decisions and so there is no moral need to eliminate differentials based upon
so-called option luck (i.e. that result from people’s deliberate and wellreasoned choices).8
Dworkin (2000) goes to great lengths to describe how one would
structure an insurance market, be it public or private, to avoid injustice
resulting from brute luck while accommodating alternative choices and
outcomes due to option luck and varying preferences (within reason). As
an example, he suggests what would be chosen in a hypothetical employment insurance scheme designed under the presumption that “everyone
assumes that each person, including himself, is equally likely to lose his
job” (2000, 332). This hypothetical veil of ignorance, similar to that of
Harsanyi (1953, 1955), would lead to a market opportunity that does not
favour the fortunate over the unfortunate. Dworkin does allow, however,
that people would have different preferences and might reasonably want
to choose somewhat different coverage, or may have rationally decided
— 132 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 132
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
to invest more or less in training in a way that affects the likelihood of
becoming unemployed. Thus, an insurance scheme that does not eliminate all inequalities is not necessarily unjust. However, differential treatment arising from brute luck differences, such as different genetic inheritance, would be unjust. Therefore, for the most part the luck egalitarian
view would support a ban on insurers using genetic test results for ratemaking purposes.
Dworkin accommodates some variation in coverage among individuals that could be justified “by asking what unemployment insurance people with a representative mixture of the tastes and ambitions most Americans have […] would buy if they had the wealth that is average among
us and were acting prudently” (2000, 333). In the context of health insurance, it is relatively straightforward to ascertain a reasonably argued minimum level of healthcare that individuals would choose from behind this
veil of ignorance. Obviously everyone would want the cost of necessary
surgeries, such as appendectomies, covered as long as these are not prohibitively expensive in terms of the effect of including them in a health
insurance plan from the perspective of the average income earner. Many
have argued that justice would be well-served by a single payer health
insurance plan, which is mandatory and affordable to all (i.e. provided to
the poor without cost and financed at least partially through taxation).
Brute luck, furthermore, such as that created by unfortunate genetic conditions, should not be reflected in the cost of insurance borne by any
individual. One may also accept that one could allow for additional
aspects of health care provision to be covered by supplementary (or alternative) private health insurance schemes accommodating heterogeneous
preferences among individuals. This raises the question whether such
private insurance should be regulated so that those with bad brute luck
(e.g. bad genes) should be buffered from being charged higher prices. At
first glance, the appropriate response would be in the affirmative; a ban
on risk-rating through the use of genetic test results seems a necessary
component of justice. However, given that the insurer is generally not
— 133 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 133
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
privy to personal preferences, it is not clear to the insurer whether an
individual who wishes to purchase additional supplementary insurance
does so because of his or her preference for more healthcare or due to
his or her risk level being higher. Thus, adverse selection may arise
unavoidably if insurance is structured to allow variations in coverage level
but prices cannot vary by risk level.
This seems to be the case even more so for life insurance. Individual
preferences or demands for life insurance vary greatly across the population and so determining an appropriate coverage level that should be
offered through a public (mandatory) insurance scheme would be problematic.9 Nonetheless, many would argue that it may well serve justice to
ban insurers’ use of genetic test results for pricing life insurance. It is
these complications that we investigate here regarding arguments based
on Luck Egalitarianism and other moral and pragmatic perspectives.
A major opponent to restrictions on the use of genetic testing is
often, of course, the insurance industry. A concept specific to the insurance industry is that price should reflect, as closely as possible, expected
cost for each policy holder – a view commonly described as actuarial
fairness. One of the insurance industry’s biggest fears is that the consumer will have access to genetic tests that are not made available to the
companies. The result would be that those consumers who discover unfavourable results from genetic tests could then buy huge quantities of
insurance in the full knowledge that their health or mortality risk will
allow them (or their heirs) to realize enormous financial benefits. In a
letter dated February 16, 2007 to then American Speaker of the House,
Nancy Pelosi, the American Academy of Actuaries, a group that advises
federal policymakers on issues related to insurance, writes: “Risk classification, which is the process by which applicants for insurance coverage
are placed into groups of roughly equivalent levels of risk to ensure their
premium cost is commensurate with their risk level, is key to the soundness of the voluntary individual medical expense insurance market.
Voluntary markets operate most efficiently when there is equality of
— 134 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 134
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
information among buyers and sellers. […] barring insurers from obtaining test results already known to the applicant could result in an imbalance of information that would leave insurers at a disadvantage. Such
asymmetric information between the insurer and the applicant could
result in adverse selection that would have a direct impact on premium
rates, ultimately raising the cost of insurance to everyone.”10
In most circumstances, economic reasoning also supports the business/actuarial view. This is the idea that one party in a contract or transaction should not be allowed to hide relevant information. The classic
paper of Akerlof (1970) explains how the presence of hidden information
may spoil the market for everyone. His example of used cars offers a
simple understanding of the inherent logic. Suppose some sellers of used
cars hold a poor quality product, a so-called lemon, while others hold a
product of good quality. Buyers are willing to pay a higher price for
a higher quality product and sellers of good quality cars typically require
a higher price to part with their vehicles. Suppose further that potential
buyers cannot discern between good or poor quality cars, but they are
aware that sellers of both types are in the market. The result is that the
price buyers are willing to pay for used cars will be based on their willingness to pay for a car of ‘average quality’, and this price will be below the
value to both buyer and seller of the high quality cars. At least some sellers of good quality cars can be expected to withdraw their cars from the
market as they would not be willing to sell their vehicles at the required
discount. This will shift the balance of used cars left in the market towards
poor quality cars and thus lead to an even lower price. Faced with an even
higher discount in order to sell a good quality car, even more good
quality cars will be withdrawn from the market. Although an equilibrium
may be reached with some cars of both quality being sold, there will likely
be many good quality cars that sellers would like to sell and potential
buyers would like to buy, but which nevertheless go unsold. Thus, the
market does not operate as efficiently as it would if the information about
product quality was not hidden. One could even end up in an equilibrium
— 135 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 135
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
where no good quality used cars are being sold – a so-called adverse
selection death cycle. This efficiency consideration is explained more fully
in the following section in the context of an insurance market when individuals are able to keep their genetic test results private.
IV. MARKET IMPLICATIONS OF REGULATORY BANS ON GENETIC
INFORMATION
A ban on insurers’ using genetic test results for pricing insurance contracts creates just such a situation of asymmetric information that actuaries predict. The buyer of the product has information relevant to the
payoff (expected cost) of the seller. This scenario has been referred to as
regulatory adverse selection.11 One strategy that may be adopted by insurers in such a scenario we will call ‘simple pooling’. An insurer who cannot
distinguish the good risks from the bad may simply pool the consumers
together and charge a price high enough to cover the costs. At any given
price, the higher risk types value the product more since they are more
likely to make a claim. Consequently, one can expect higher risk types to
purchase more insurance than lower risk types.12 Thus, the price that is
required to cover the pooled actuarial cost of the insurance sold will be
higher than the simple population weighted average cost. Such a price is
referred to as the average clientele risk price.13 This price can still be referred
to as actuarially fair, but it is a pooled actuarially fair price rather than a
risk type specific actuarially fair price. Also, the price will typically be
between the price that would be charged for high and low risk types and
so the suggestion that is sometimes made that allowing consumers to
keep such information private will lead to an increase in the price of
insurance for everyone is not credible.14 It is most likely, however, that a
ban would lead to a price of insurance that is higher than the average
(population weighted) price that would be charged in the absence of such
a ban and certainly higher than what the low risk types would pay. This
— 136 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 136
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
is the reason low risk types purchase less insurance under ban conditions,
reflecting the same type of market inefficiency associated with the example in which the average price for low and high quality cars is such that
sellers of high quality cars are less likely to sell their cars due to the discount generated by privately held information. Adverse selection restricts
market transactions that would be welfare improving. This is the primary
concern about such a ban from an economic efficiency perspective. The
social importance of life insurance is reflected in the study by Bernheim
et al. who estimate that “35.7 percent of poverty among surviving women
[…] resulted from a failure to ensure an undiminished living standard
through insurance” (2003, 361). Nonetheless, one must balance the concern of under-insurance of low risk types due to a ban on insurers using
genetic information with the issue of the price and availability of insurance for the bad (high) genetic risks. We address this in the following
section.
A second strategy that insurers may follow rather than simple pooling
is to design different contracts that appeal to a greater or lesser extent to
different risk types. For example, since higher risk individuals place a
higher value on more comprehensive coverage, it follows that insurers
can offer one policy with a high level of coverage at a relatively high unit
price and another contract with a lower level of coverage at a lower unit
price. The bad or high risk type is more willing to pay the higher unit price
for the higher coverage policy and as a result such a strategy can lead to
a so-called ‘self selection equilibrium’, with high risk types purchasing
the more costly policy with more comprehensive coverage while low
risk types purchase the less costly policy with a lower level of coverage.
This is called a ‘separating contracts outcome’. Rothschild and Stiglitz
(1976) and Wilson (1977) showed that if the fraction of high risk types
in the population is sufficiently high, then such an outcome can be
expected and that the high risk types are no better-off than if the insurer
were allowed to use the privately held information (i.e. the genetic test
results). The low risk types, however, end up with an inefficiently low
— 137 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 137
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
level of coverage and so the overall result of the ban is to harm good
(low) risks without providing any advantage to bad (high) risks. This is a
compelling efficiency argument against banning insurers from using
genetic test results. However, this market outcome is a credible scenario
only if the proportion of high risk types in the population is high enough
that low risk types prefer their contract (from the separating pair of contracts), which has limited coverage, to a pooling contract with higher
coverage but which is also offered at a higher price.15
For a separating contracts outcome to be effective, it is also necessary
that the insurers offer exclusive contracts; that is, each consumer can purchase insurance from only one of the possible sellers. Without this restriction, the high risk types can purchase from several insurers the contract
with the low coverage and low price level that is offered with the expectation of attracting only low risk types. The result is that each insurer will
discover that the contract designed for the low risk types will attract many
high risk customers and so the low price will not be sustainable. Although
some insurance products are sold through exclusive contracts (e.g. automobile, health, long term care), life insurance typically is not.16 So the
separating contracts scenario does not apply to all types of insurance.
We focus here on two informational scenarios. In the first, genetic
information is available to clients who can use it to make choices regarding their insurance coverage, but insurers are banned from using it to
risk-rate premiums. In the second, the information can be acquired by
clients but, if they do, insurers have access to the information and can
use it for pricing decisions. We consider these scenarios in the context of
a free market for insurance, as is typically the case for life insurance provision. In such markets individuals can use private information about
their mortality risk strategically in deciding how much insurance to purchase. This is the context which interests us most. Health insurance, on
the other hand, is provided publicly in many countries (at least for basic
healthcare services). In a country where health insurance is mandatory but
provided by private companies, as it is in Switzerland, clients cannot use
— 138 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 138
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
genetic information to choose strategically their level of insurance coverage within the basic package. However, as in Switzerland, it is often the
case that supplemental insurance is optional and various levels of a
deductible can be chosen. In such economic environments the range of
basic items covered is typically mandated and insurers are not allowed to
use information, such as genetic test results, to charge higher risk individuals a higher premium. In order to discourage insurers from trying to
attract a disproportionate share of low risk clients (i.e. ‘cream-skimming’)
governments often offer subsidies to insurers for their clients who
possess high risk characteristics. This process is known as ‘risk-adjustment’. The implications of a ban on insurers using genetic test results to
risk-rate the insured in such a setting will depend on many factors, including how comprehensive is the mandated coverage and whether effective
risk-adjustment is offered (Van de Ven and Ellis 2000). In Australia, as
in some other countries, private health insurers coexist with public insurance and the private insurers are restricted in the information they can
use to risk-rate clients. This has lead to substantial adverse selection
effects with private (enhanced) coverage being relatively more attractive
to high risk types and low risk types tending to migrate to the public plan
(Butler 2003; Connelly et al. 2010). The range of complications associated
with restrictions to insurers’ use of genetic test results, or to any type of
risk relevant information, depends on many market features that vary
widely across countries in which private health insurance plans coexist
with public plans or are sponsored at least partially by government.
A comprehensive analysis of these possibilities is beyond the scope of the
present contribution.
V. WELFARIST PERSPECTIVE
As well as providing an understanding of the possible market implications
of banning the use of genetic test results, economic analysis can also offer
— 139 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 139
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
an assessment based on standard welfare criteria of efficiency and equity.
The efficiency implications of a ban have already been described in the
previous section, although only from an interim temporal perspective.
From this perspective (i.e. after we know our genetic type and have made
our insurance purchase but before we know our claim status) insurance
is treated as an individualistic risk management tool. In the context of a
pooling equilibrium, some people – the low risk types – end up with too
little insurance while others – the high risk types – end up with too much
insurance. If the quantitative relevance of the genetic information held
privately by consumers is sufficiently great, then one can expect the extent
of under-insurance (of good risks) to be significant enough that the ban
should be judged to be undesirable.17 This is true a fortiori if the result in
the market is one of separating contracts (i.e. a high coverage policy at a
high price for the bad risks and a low coverage policy at a low price for
the good risks), since high risk types are no better-off as a result of the
regulatory ban, while low risk types are made worse-off.
If the market reaction to a ban is a simple pooling contract then,
compared to no ban, high risk types face a lower price and low risk types
face a higher price; that is, there is cross-subsidization from the lucky (low
risks) to the unlucky (high risks). Comparing incomes of the two types
net of the cost of buying insurance, the ban (implicitly) creates a progressive transfer from better-off to less well-off. If one adopts a general
welfarist framework, which spans a range of distributive ethics from utilitarian to Rawlsian (lexicographic maximin), then one may indeed find
support for a ban if the extent of adverse selection is small enough that
underinsurance of low risk types is sufficiently unimportant in comparison to the implicit (favourable) redistribution from good risks to bad
risks. Hoy (2006) has shown that for any welfare function reflecting at
least some modest aversion to inequality, then for any amount of information below some critical level, welfare is enhanced by the presence of
a ban. Although many parameters play a role in leading to this outcome,
such as the extent to which people are risk averse, ceteris paribus the
— 140 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 140
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
critical level is characterized by the fraction of individuals in the population who are of high risk.
Therefore, adopting a welfare perspective allows one to form a coherent basis on whether a regulatory ban on the use of genetic test results is
desirable. The difficulty with applying such an approach, however, is that
at one point in time (e.g. now) the fraction of people who know they are
of high risk may be small enough that a welfare criterion supports a ban,
while at another point in time (e.g. 20 years from now) the state of information may be such that the recommended policy conclusion will be
reversed. This does not imply an inconsistency in the assessment principle
per se but does mean the application may lead to different recommendations depending on circumstances. Certainly one would not suggest an
irrevocable ban on insurers using genetic test results for pricing purposes
as a result of a standard economic welfare assessment.
VI. THE MORAL DISCRIMINATION PERSPECTIVE
Whether one believes a regulatory ban is justified by a concern with unfair
discrimination may well depend on the type of equilibrium, pooling or
separating, as well as the temporal perspective one deems most relevant.
Furthermore, it must be decided whether one is concerned with price
discrimination, quantity discrimination or some other perspective such as
‘well-being’ discrimination.
Consider first the distinction between price and quantity discrimination. In the event that a regulatory ban results in a pooling equilibrium,
both risk types face the same price and so price discrimination is absent
(at least from a morality-based view of discrimination, since brute luck
is not allowed to have price implications). One could argue that from
both ex ante and ex post temporal perspectives no price discrimination
occurs. Low risk types, however, choose to purchase less insurance and
so it could be argued that they face discrimination along the quantity
— 141 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 141
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
dimension. Such a stance is perhaps not without controversy, however,
since the level of coverage is a freely engaged choice (i.e. a result of
option luck). In the context of life insurance, however, the affected party
or group of relevance is the survivor family and this is a different entity
from the individual or ‘unit’ who/which made the decision about how
much insurance to purchase in the first place. It is clear that survivor
families of low risk types are, ceteris paribus, worse-off in a strictly economic sense than are survivor families of high risk types as the latter will
hold larger quantities of life insurance. Thus, from an ex post perspective,
the ‘quantity discrimination’ experienced by low risk types is a relevant
concern. This experience can also be identified as a sort of (ex post) ‘wellbeing’ discrimination against survivor families of low risk types.18
In the case of a regulatory ban leading to a separating equilibrium,
the low risk types again end up with a lower level of insurance coverage and so face quantity discrimination and the survivor families face
well-being discrimination. An added complication arises due to the fact
that the insured face a pair of contracts – one with a higher price but
higher quantity that attracts high risk types and the other with a lower
price but lower quantity that attracts low risk types. From an interim
temporal perspective (i.e. at the point in time when a decision about
which insurance contract to choose is to be taken) both types are free
to purchase either policy and so one might argue that there is no discrimination. However, the high risk types purchase the high price, high
quantity contract while the low risk types purchase the low price, low
quantity contract. So, leaving the relevance of free choice aside, the
high risk types are discriminated against in price while the low risk
types are discriminated against in the quantity dimension. It is not clear
which type of discrimination is more relevant in the context of the
interim temporal perspective. From an ex post perspective, however, it
is clear that the survivor families of low risk types again end up worseoff as a result of a regulatory ban and so can be said to face ‘well-being’
discrimination.
— 142 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 142
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
VII. THE BUSINESS/ACTUARIAL PERSPECTIVE
As noted above, the insurance industry argues that insurance companies
should be allowed to use actuarially relevant genetic test results held by
the insured. On the face of it at least, this is a reasonable position. Contracts in private sector relations are typically expected to be entered into
in good faith; that is, any relevant information that either party holds
should not be kept hidden. Moreover, since the main cost of providing
insurance is generated from satisfying claims and the expected cost of
doing so for high risk types is indeed higher, it is reasonable to argue that
in a pragmatic sense the insurer is not engaged in unfair discrimination
when high risk types are assessed a higher unit price for insurance coverage. After all, a product that costs more to supply is typically sold at a
higher price. This seems a reasonable way for a supplier to behave. On
the other hand, however, private sector interactions are often subject to
regulations that are deemed to be in the public interest. It can be argued
with reason that private incentives should be allowed to play out in the
market place only if the public interest is served. What’s more, it is generally the case that insurers do not have perfect information about each
individual and so all individuals are placed into groups in which some
differences in risk levels are present. Introducing additional information,
such as the result of a genetic test, is also likely to be merely an imperfect
indication of an individual’s overall true risk type. As a result, the use of
genetic tests enhances information but does not lead to perfect risk-type
specific categorization. Hoy and Lambert (2000) have shown that information that improves the average accuracy of risk determination may
actually create an additional impact of discrimination of the statistical
(rather than moral) sort if one allows for the realization that the cost of
discrimination may indeed be nonlinear in the amount.19 As long as all
insurers face the same regulation, then prices can be expected to adjust
in such a way as to preserve the financial viability of insurance companies
even if some relevant information about individual risk levels is not used.
— 143 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 143
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
Thus, the claim that insurers should or must use all available information
to classify individuals is neither true descriptively nor justifiable on normative grounds. Of course, if a sufficiently large amount of information
about risk type can be held privately by the insured, then it is quite possible that the market will not function very well as a means of risk management for them. It may thus not be sensible to adopt a regulation that
suppresses such information in all circumstances.
VIII. FURTHER CONSIDERATIONS
We have discussed many complexities – both philosophical and practical
– that arise when addressing the implications and desirability of a regulatory ban on insurers’ use of genetic test results for pricing insurance
contracts in the context of a free insurance market. In this section, we
explore a couple of additional considerations that the future may hold
and that will add further challenges to the development of a coherent
framework for deciding on the efficacy of a ban.
Although early discoveries in the way genes are linked to diseases
tended mostly to be for Mendelian (monogenic) diseases, current and
future research is more likely to lead to the discovery of more multi-factorial disease genes that involve more than one gene in combination with
environmental factors (Antonarakis and Bechmann 2006). In many cases,
these environmental factors can be at least partially controlled through
individuals’ behaviour. Such genes are sometimes referred to as predisposition genes since their presence leads to an increased likelihood of a person contracting a specific disease (e.g. lung cancer). Individual behaviours
can influence this risk (e.g. by avoiding smoking) with differential impact
on the risk of the onset of a disease depending on which genes are present.
Thus, whether or not a regulatory ban is in place may influence a person’s
behaviour about environmental exposures or lifestyle decisions as well as
insurers’ pricing strategies. Consider, for example, that a regulatory ban is
— 144 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 144
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
not in place and so insurers can charge a higher price to anyone known to
have a higher mortality risk due to their having some predisposition gene.
Since individuals have some control over their risk level, it is less clear how
one should view price differentials based on individuals possessing such a
gene. To some extent, risk differences are the result of individual choices
that could come under the option luck category. Whether or not the
insurer can observe or determine the individual behaviours and use this
information in determining individual specific prices comes into play when
assessing whether unfair discrimination is being practised. One might
argue that, to the extent that risk levels differ because of differences in
behaviour (i.e. option luck), insurers should be allowed to assign different
prices. However, separating that part of differences in risk levels associated
with option luck (behaviour differences) and that part due to brute luck
(genetic differences) is not a trivial exercise. Therefore, it is a difficult challenge to determine whether differential prices are justifiable or what proportion of differences in risk levels it would be justifiable to reflect in
prices. Welfare analysis is also more complex when differences are due to
multi-factorial genes (Hoy 1989).20
The entire debate about appropriate regulation may shift as the costs
of genetic tests fall over time and the market for direct-to-consumer
genetic testing (DTC-GT) develops. Over the counter tests have already
raised grave concerns about genetic information being offered to a public
not sufficiently educated to interpret it or not always aware of the importance of protecting the information. Although the companies concerned
(e.g. www.23andme.com and www.navigenics.com) include a statement
about security and privacy, nothing sent over the internet has 100% assurance of protection of personal information. The companies invite clients
to participate in research, including completing questionnaires that gather
personal data beyond the genetic test itself, and in so doing increasingly
expand the scope for loss of personal privacy.
If genetic tests become sufficiently cheap and insurers become
concerned that many potential customers will have substantial genetic
— 145 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 145
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
information obtained from the DTC-GT market, then insurers may well
want to screen all applicants by insisting that they take a battery of genetic
tests in order to qualify for insurance. However, outside the need to
qualify for insurance, many people do not want to be tested in this manner.21 Not having a regulation banning insurers from insisting on genetic
testing means another right will be placed at risk; namely, the right not
to know about one’s genetic status (Lemmens, Luther and Hoy 2008).22
Throughout our contribution, the quantity of relevant genetic information that is held by the insurance buying public has been the most
crucial feature of the contracting environment and it can tip the balance
on just how adverse selection will be realized in the market place. This in
turn has a fundamental affect on the way any of the criteria considered
here performs as a useful framework for deciding on the merits of a
regulatory ban. In a review of the current actuarial (academic) literature
MacDonald concludes that “little, if any strong empirical evidence has
been found for the presence of adverse selection (although it is admittedly
hard to study)” (2009, 4). Simulation exercises based on population genetics and epidemiological data (Hoy et al. 2003; Hoy and Witt 2007) also
find that, for the most part, little impact is likely to occur from a ban on
insurers using genetic test results for health and life insurance markets,
respectively. Oster et al. (2010), however, report strong evidence of
adverse selection in the long-term care insurance market due to individuals holding private information about Huntington disease carrier status.
It seems safe to say that the future holds uncertainty in this regard and
continued empirical research will be necessary in order to help resolve the
debate about the use of genetic information in insurance markets.
IX. CONCLUSION
There may be good reasons why different countries have different regulations regarding phenomena such as the use of genetic information by
— 146 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 146
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
insurers. However, it is at least curious that countries with similar protections against racial, religious, and gender discrimination choose such
different regulations in regards to genetic discrimination. Even within the
OECD countries, the response to the concern of genetic privacy and
genetic discrimination has varied from no regulation whatsoever to rigid
and complete bans on the use of genetic information by insurers. We
argue that this is not surprising, given that the academic literature has not
provided a unified or coherent assessment of the debate. This, we argue,
is due to several different sets of issues that make such an exercise
extremely challenging. The various elements that form any such analysis
interact in a complex manner. First, whether the contracting environment
involves exclusivity or non-exclusivity of provision determines whether
pooling or separating contracts will persist under a ban. This in turn
affects the economic implications of each of the assessments of a ban
that we have considered. Second, the reasonable normative and other
criteria we have reviewed that can be used to develop guidelines or principles for regulation often lead to starkly different conclusions across the
various contracting environments.
Third, the conclusions about regulation that would be recommended
using any of the criteria presented in this article and reflected in the
literature depend critically on the temporal perspective one adopts. Should
one perform an evaluation of a ban from the perspective of individuals
before they know their risk type? Currently, most people have limited or
no information about their genetic risk profile, but this will certainly
change in the future. Thus, this temporal perspective (i.e. the ex ante position) seems a perfectly reasonable one to adopt in the context of genetic
discrimination. However, once people know their risk type, their private
incentives and that of insurance providers change. It is this stage (i.e. the
interim stage) in which customers and insurance firms engage in contracting. It also seems quite a reasonable point from which to assess the merits of regulating these market transactions. Finally, since the point of
insurance is protection of individuals in the event of a financial loss, then
— 147 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 147
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
an ex post perspective, wherein all uncertainty about the relevant states of
the world has been resolved, may seem most appropriate. This is particularly relevant in the context of life insurance, since it is the financial
protection of survivor families that is the goal of insurance and the survivor family is a different entity from the decision making unit in the ex
ante or interim stages.
We find that even within the same analytic framework one can arrive
at different conclusions regarding the evaluation of the merits of a ban.
This is especially so for the approach we have associated with a moral
definition of discrimination. One reason that this normative perspective
faces such complexities is that it is not clear what element should be the
main focus of concern. Should it be price discrimination, quantity discrimination, or end state discrimination in well-being? It is possible that,
from an ex ante perspective, one would conclude that no discrimination
occurs in the presence of a ban, but from an interim or ex post scenario
there is one form of discrimination faced by one risk type (e.g. price discrimination against high risk types) and another form of discrimination
faced by the other risk type (e.g. quantity discrimination or well-being
discrimination against low risk types). That such complexities in reasoning
arise for a viewpoint that at face value is quite straightforward demonstrates the challenges involved in arriving at a coherent and normatively
appealing policy.
If there is a main message to offer in regards to how to resolve or at
least balance the complexities exposed in the present contribution, it is
that, regardless of the value judgements one believes most ethically pleasing, a careful consideration of how the economic implications of a ban
on insurers’ use of genetic test results to price insurance products will play
out should be an important element in any argument in favour or against
such regulation. Even if one feels it is important not to treat anyone differently for reasons that arise from circumstances outside of the individual’s control, as do luck egalitarians, banning insurers’ use of genetic
test results may lead to ex post inefficiencies in the insurance market
— 148 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 148
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
including substantial underinsurance for those who, ex ante, are genetically
lucky. In such circumstances, the severe inefficiency implications can reasonably be argued to dominate concerns of fairness for the genetically
unlucky. On the other hand, if the extent of inefficiency induced by a ban
is low, as may be the case for example with a pooling equilibrium outcome, then the equality enhancement resulting from a ban on the use of
genetic tests by insurers is appealing from a variety of equity-based considerations including the right to privacy, Luck Egalitarianism, social welfare, and unfair discrimination.23
WORKS CITED
Akerlof, George. A. 1970. “The Market for Lemons: Quality Uncertainty and the
Market Mechanism.” Quarterly Journal of Economics 84/3: 488-500.
Alper, Joseph S. and Marvin R. Natowicz. 1993. “Genetic Testing and Insurance.”
British Medical Journal 307: 1506-1507.
Antonarakis, Stylianos E. and Jacques S. Bechmann. 2006. “Mendelian Disorders
Deserve More Attention.” Nature Reviews Genetics 7: 277-282.
Babul, Riyan, Shelin Adam, Berry Kremer, Suzanne Dufrasne, Sandi Wiggins, Marlene
Huggins, Jane Theilmann, Maurice Bloch and Michael R. Hayden. 1993. “Attitudes
Towards Direct Predictive Testing for the Huntington’s Disease Gene.” Journal of
the American Medical Association 270: 2321-2325.
Bernheim, B. Douglas, Lorenzo Forni, Jagadeesh Gokhale and Laurence J. Kotlifkoff.
2003. “The Mismatch between Life Insurance Holdings and Financial Vulnerabilities: Evidence from the Health and Retirement Study.” American Economic Review
93/1: 354-65.
Butler, James R. G. 2003. “Adverse Selection, Genetic Testing and Life Insurance – Lessons from Health Insurance in Australia.” Agenda 10/1: 73-89.
Connelly, Luke B., Francesco Paolucci, James R. G. Butler and Paul Collins. 2010. “Risk
Equalisation and Voluntary Health Insurance Markets: The Case of Australia.”
Health Policy 98/1: 3-14.
Crocker, Keith J. and Arthur Snow. 1986. “The Efficiency Effects of Categorical Discrimination in the Insurance Industry.” The Journal of Political Economy 94/2: 321-344.
Daniels, Norman. 1994. “The Articulation of Values and Principles Involved in Health
Care Reform.” The Journal of Medicine and Philosophy 19/5: 425-433.
Diamond, Peter A. 1967. “Cardinal Welfare, Individualistic Ethics, and Interpersonal
Comparison of Utility: Comment.” The Journal of Political Economy 75/5: 765-766.
— 149 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 149
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
Dworkin, Ronald. 2000. Sovereign Virtue: The Theory and Practice of Equality. Cambridge,
MA: Harvard University Press.
Filipova, Lilia and Michael Hoy. 2009. “Impact of Genetic Testing on Surveillance and
Prevention.” Department of Economics Working Paper no. 2009-4. University of
Guelph.
Fleurbaey, Marc. 2010. “Assessing Risky Social Situations.” Journal of Political Economy
118/4: 649-680.
Hall, Mark A., Jean E. McEwen, James C. Barton, Ann P. Walker, Edmund G. Howe,
Jacob A. Reiss, Tara E. Power, Shellie D. Ellis, Diane C. Tucker, Barbara W. Harrison, Gordon D. McLaren, Andrea Ruggiero and Elizabeth J. Thomson. 2005.
“Concerns in Primary Care Population about Genetic Discrimination by Insurers.”
Genetics in Medicine 7/5: 311-316.
Hammond, Peter. 1981. “Ex-ante and Ex-Post Welfare Optimality under Uncertainty.”
Economica 48/191: 235-250.
Harsanyi, John C. 1953. “Cardinal Utility in Welfare Economics and in the Theory of
Risk Taking.” Journal of Political Economy 61/5: 434-435.
Harsanyi, John C. 1955. “Cardinal Welfare, Individualistic Ethics, and Interpersonal
Comparisons of Utility.” Journal of Political Economy 63/4: 309-321.
Hoy, Michael. 1982. “Categorizing Risks in the Insurance Industry.” Quarterly Journal of
Economics 97/2: 321-336.
Hoy, Michael. 1989. “The Value of Screening Mechanisms under Alternative Insurance
Possibilities.” Journal of Public Economics 39/2: 177-206.
Hoy, Michael and Mattias K. Polborn. 2000. “The Value of Genetic Information in the
Life Insurance Market.” Journal of Public Economics 78/3: 235-252.
Hoy, Michael and Peter J. Lambert. 2000. “Genetic Screening and Price Discrimination
in Insurance Markets.” The Geneva Papers on Risk and Insurance Theory 25/2: 103-130.
Hoy, Michael, Fabienne Orsi, Francois Eisinger and Jean-Paul Moatti. 2003. “The
Impact of Genetic Testing on Health Care Insurance.” Geneva Papers on Risk and
Insurance: Issues and Practice 28/2: 203-221.
Hoy, Michael and Michael Ruse. 2005. “Regulating Genetic Information in Insurance
Markets.” Risk Management and Insurance Review 8/2: 211-237.
Hoy, Michael. 2006. “Risk Classification and Social Welfare.” The Geneva Papers on Risk
and Insurance: Issues and Practice 31/2: 245-269.
Hoy, Michael and Julia Witt. 2007. “Welfare Effects of Banning Genetic Information in
the Life Insurance Market: The Case of BRCA1/2 Genes.” Journal of Risk and Insurance 74/3: 523-546.
Joly, Yann, Maria Braker and Michael Le Huynh. 2010. “Genetic Discrimination in
Private Insurance: Global Perspectives.” New Genetics in Society 29/4: 351-368
Kitcher, Philip. 1996. The Lives to Come: The Genetic Revolution and Human Possibilities. New
York: Simon & Schuster.
Klitzman, Robert. 2010. “Views of Discrimination among Individuals Confronting
Genetic Disease.” Journal of Genetic Counseling 19: 68-83.
— 150 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 150
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
Lapham, E. Virginia, Chahira Kozma and Joan O. Weiss. 1996. “Genetic Discrimination: Perspectives of Consumers.” Science 274: 621-623.
Ledyard, John O. and Thomas R. Palfrey. 1999. “A Characterization of Interim Efficiency with Public Goods.” Econometrica 67/2: 435-448.
Lemmens, Trudo, Lori Luther and Michael Hoy. 2008. “Genetic Information Access, a
Legal Perspective: A Duty to Know or a Right Not to Know, and a Duty or Option
to Warn?” In Encyclopedia of Life Sciences (ELS), 1-5. Chichester: John Wiley & Sons.
Macdonald, Angus S. 2009, “Genetic Factors in Life Insurance: Actuarial Basis,” Encyclopedia of Life Sciences (ELS), 1-5. Chichester: John Wiley & Sons.
Martins-da-Rocha, V. Filipe. 2010. “Interim Efficiency with MEU-preferences.” Journal
of Economic Theory 145/5: 1987-2017.
Meiser, Bettina and Stewart Dunn. 2000. “Psychological Impact of Genetic Testing for
Huntington’s Disease: An Update of the Literature.” Journal of Neurology, Neurosurgery
and Psychiatry 69: 574.
Nelkin, Dorothy and M. Susan Lindee. 1995. The DNA Mystique. New York: Freeman.
Nordgren, Anders and Eric T. Juengst. 2009. “Can Genomics Tell Me Who I Am?
Essentialist Rhetoric in Direct-to-consumer DNA Testing.” New Genetics in Society
28/2: 157-172.
O’Neil, Martin. 2006. “Genetic Information, Life Insurance, and Social Justice.” The
Monist. 89/4: 567-592.
Oster, Emily, Ira Shoulson, Kimberly Quaid and E. Ray Dorsey. 2010. “Genetic Adverse
Selection: Evidence from Long-Term Care Insurance and Huntington Disease.”
Journal of Public Economics 94/11-12: 1041-1050.
Pauly, Mark V., Kate H. Withers, Krupa Subramanian-Viswanathan, Jean Lemaire, John
C. Hershey, Katrina Armstrong and David A. Asch. 2003. “Price Elasticity of
Demand for Term Life Insurance and Adverse Selection.” National Bureau of
Economic Research Working Paper #9925. Cambridge, MA: NBER.
Polborn, Mattias K., Michael Hoy and Asha Sadanand. 2006. “Advantageous Effects of
Regulatory Adverse Selection in the Life Insurance Market.” The Economic Journal
116/508: 327-354.
Pollara-Earnscliffe. 2003. Public Opinion Research into Genetic Privacy Issues: Final Report,
Government of Canada. http://www.bioportal.gc.ca/CMFiles/geneticprivacypormar2003E49 OCT-912004-6986.pdf [accessed January 13, 2012].
Quaid, Kimberly A. and Michael Morris. 1993. “Reluctance to Undergo Genetic Testing:
The Case of Huntington Disease.” American Journal of Medical Genetics 45/1: 4145.
Rothschild, Michael and Joseph Stiglitz. 1976. “Equilibrium in Competitive Insurance
Markets: An Essay on the Economics of Imperfect Information.” Quarterly Journal
of Economics 90/4: 630-649.
Rothstein, Mark A. and Carlton A. Hornung. 2004. “Public Attitudes.” In Genetics and
Life Insurance: Medical Underwriting and Social Policy. Edited by Mark A. Rothstein,
1-25. Cambridge, MA: MIT Press.
— 151 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 151
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
Sen, Amartya. 2002. “Why Health Equity.” Health Economics 11/8: 659-666.
Segall, Shlomi. 2010. Health, Luck, and Justice. Princeton: Princeton University Press.
Thomas, R. Guy. 2008. “Loss Coverage as a Public Policy Objective for Risk Classification Schemes.” Journal of Risk and Insurance 75/4: 997-1018.
Van de Ven, Wynand P. and Randall P. Ellis. 2000. “Risk Adjustment in Competitive
Health Plan Markets.” In Handbook of Health Economics Vol. 1. Edited by Anthony
Culyer and Joseph Newhouse, 756-845. Amsterdam: Elsevier.
Wilson, Charles. 1977. “A Model of Insurance with Incomplete Information.” Journal of
Economic Theory 16/2: 167-207.
Wolpe, Paul R. 1997. “If I Am Only My Genes, What Am I? Genetic Essentialism and
a Jewish Response.” Kennedy Institute of Ethics Journal 7/3: 213-230.
NOTES
1. This has been an important consideration in the debate about the normative foundations
of welfare economics and the appropriate choice of temporal perspective for assessing individual
well-being. See, for example, Diamond (1967), Hammond (1981), Ledyard and Palfrey (1999),
Martins-da-Rocha (2010) and Fleurbaey (2010).
2. See Joly, Braker, Le Huynh (2010) for details.
3. A Canadian poll conducted in 2003 by Pollara-Earnscliffe found that 91% of respondents
believe insurance companies should not have the right to ask about an individual’s genetic information.
4. See the website http://www.ccgf-cceg.ca/en/about-ccgf#members for a list of the members
of the Canadian Coalition for Genetic Fairness and for a discussion of their principles.
5. Lapham, Kozma and Weiss (1996) report that in a study of the perceptions of 332
members of genetic support groups that 9% refused genetic testing due to fear of discrimination
by insurance companies or (potential) employers. Hall, McEwan, Barton, Walker et al. (2005)
found in a survey of 86,859 adults in primary-care screening for hereditary hemochromatosis (iron
overload) that 40% agreed that “Genetic testing is not a good idea because you might have
trouble getting or keeping your insurance.” Klitzman (2010) surveys the literature regarding views
of genetic discrimination.
6. For further discussion, see also Babul et al. (1993) and Quaid and Morris (1993).
7. This is a figure that corresponds to the percentage of people who actually purchase life
insurance policies.
8. Canonically, Ronald Dworkin explains option luck as follows: “Option luck is a matter
of how deliberate and calculated gambles turn out – whether someone gains or loses through
accepting an isolated risk he or she should have anticipated and might have declined” (2000, 73).
Brute luck is “a matter of how risks fall out that are not in that sense deliberate gambles” (2000,
73). If I suddenly go blind as a result of a genetic condition, my brute luck is bad, but if I buy a
— 152 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 152
16/03/12 09:43
M . DURNIN , M . HOY AND M . RUSE – GENETIC TESTING AND INSURANCE
lottery ticket and win, my option luck is good” (see Stanford Encyclopedia of Philosophy, http://
plato.stanford.edu /entries/justice-bad-luck/#7).
9. See Polborn, Hoy and Sadanand (2006) for a discussion of how differing preferences for
level of cover interacts with varying hidden information about risk level in designing effective
regulations for life insurance provision.
10. See www.actuary.org/pdf/health/genhouse_feb07.pdf.
11. For details of how such a ban can affect the markets for health, life and other forms of
insurance, see Hoy and Ruse (2005). The classic papers on how asymmetric information affects
insurance markets are Rothschild and Stiglitz (1976) and Wilson (1977). The market implications
when information is imperfect were first addressed in Hoy (1982) and Crocker and Snow (1986).
12. Pauly et al. (2003) estimate the risk elasticity of demand for life insurance to be between
16% and 29%. This implies, for example, that a person with a 50% higher risk of death would
purchase between 8% and 14.5% more life insurance.
13. See Hoy and Polborn (2000) for details on how such pricing for a pooling equilibrium
would arise.
14. Recall such a claim quoted earlier from the letter of the American Academy of Actuaries.
15. Determining what this critical proportion of high risk types would be depends on many
factors, including individuals’ degree of risk aversion. An application that provides some insight
into this aspect is Hoy et al. (2003).
16. Of course, if the sellers of life insurance discover over time that there is a sufficient
amount of hidden information arising from privately held genetic test results, then they may find
that exclusive contracts are worthwhile.
17. If a simple pooling contract prevails in the market, then low risks end up with less
insurance and high risks end up with more insurance compared to no ban. Thomas (2008) points
out, however, that the overall result can have more people who face a loss ending up being compensated through the claim process since the high risks are more likely to incur a loss. Although
the interim efficiency of the insurance market is still compromised by a ban, one can argue that
ex post (i.e. after all uncertainty is resolved) the fact that more losses are covered is desirable.
18. We ignore all other sources of income inequality and so consider only effects from
insurance purchases on the survivor families’ economic well-being. If high risk types experience
lower incomes during their lifetime due to their high-risk status (and possible ill health) then this
must also be weighed into the analysis.
19. That is, nonlinearity (convexity) follows if a person feels more than twice as aggrieved
by being charged 10% more than is reflected by his or her true risk level than he or she would by
being over-charged by 5%.
20. Hoy (1989) models the implications of alternative informational environments when
individuals can influence their risk level.
21. See Meiser and Dunn (2000), Babul et al. (1993) and Quaid and Morris (1993) who
document that free and anonymous genetic tests for Huntington disease are often refused by
members of at-risk families.
22. See III above for a discussion on this topic.
— 153 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 153
16/03/12 09:43
ETHICAL PERSPECTIVES – MARCH
2012
23. We thank the Fondation Brocher for accommodation and financial support for Maureen
Durnin and Michael Hoy (visiting researcher). We are also grateful for the useful comments
received from participants at the Fondation Brocher Alumni Conference in July, 2011, from
Michele Loi and from two anonymous referees.
— 154 —
Ethical Perspectives 19 (2012) 1
95288_EthPersp_2012/1_07_Durnin.indd 154
16/03/12 09:43