Insurance is pervasive in many social settings. As a cooperative device based on risk pooling, it serves to attenuate the adverse consequences of various risks (health, unemployment, natural catastrophes and so forth) by offering policyholders coverage against the losses implied by adverse events in exchange for the payment of premiums. In the insurance industry, the concept of actuarial fairness serves to establish what could be adequate, fair premiums. Accordingly, premiums paid by policyholders should match as closely as possible their risk exposure (i.e. their expected losses). Such premiums are the product of the probabilities of losses and the expected losses. This article presents a discussion of the fairness of actuarial fairness through three steps: (1) defining the concept based on its formulation within the insurance industry; (2) determining in which sense it may be about fairness; and (3) raising some objections to the actual fairness of actuarial fairness. The necessity of a normative evaluation of actuarial fairness is justified by the influence of the concept on the current reforms of public insurance systems and the fact that it highlights the question of the repartition of the gains and burdens of social cooperation.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
Tax calculation will be finalised during checkout.
Any discussion of insurance should distinguish between insurance as the general cooperative arrangement among individuals based on risk pooling and specific arrangements where an insurer acts as an intermediary between policyholders. In other words, a distinction should be made between the concept and different conceptions of insurance. This article is mainly about the concept of insurance as a cooperative mechanism. Even if mentions are made of cases where an insurer acts as an intermediary, the locus of the article remains the philosophy of insurance, i.e. the normative principles that should prevail when individuals decide to pool their risks in the face of uncertainty. We are grateful to an anonymous referee who brought this point to our attention.
Actuarial fairness as a topic of philosophical and normative enquiry is relatively underdeveloped, especially compared to fairness in redistributive justice, for instance. No matter how emerging, the ground-breaking works of various authors should be acknowledged here, in particular for their contribution to the analysis of insurance, underwriting and actuarial fairness (e.g. Baker 1996, 2000; Baker and Simon 2002; Daniels 1990; Heath 2007; Lethonen and Liukko 2011; Radetzki et al. 2008; Stone 1999–2000).
This echoes Joseph Heath’s discussion of the scope of political justice (Heath 2006b).
Underwriting is the ‘process of examining, accepting, or rejecting insurance risks, and classifying those selected, in order to charge the proper premium for each. The purpose of underwriting is to spread the risk among a pool of insureds in a manner that is equitable for the insureds and profitable for the insurer’ (Rubin 2008, p. 536) (for a detailed discussion refer to (Anonymized reference)).
‘The expected value of an act is the sum of the products of such an act (utilities × probabilities)’ (Hacking 2001, p. 80).
An externality—or external effect—is a cost (negative externality) or a gain (positive externality) that is suffered (enjoyed) by a third party to an original exchange. This initial and voluntary exchange between two individuals generates an effect that is ‘external’ in the sense that it is not part of the contract that the third party has not agreed to and that it is not accounted for in the pricing system. In this respect, externalities are cases of market failures (Papandreou 1994).
A credit default swap is ‘a credit derivative structured as a swap. One party is a lender facing a credit risk from a third party and the counterparty in the swap agrees to insure this risk in exchange for regular periodic payments (essentially an insurance premium). If the third party defaults, then the counterparty insurer will have to purchase from the insured defaulted asset. In turn, the insurer pays the insured the remaining interest on the debt as well as the principal’ (Bennett 2004, p. 84).
Risk-shifting (or risk-trading) and risk-pooling mechanisms are two different forms of risk management (Heath 2006a, pp. 323–324; Moss 2002, pp. 92–94). Risk-shifting arrangements, exemplified by CDS, consist in the transfer of risk from one agent to another, while in risk-pooling devices (such as insurance), there is no transfer of the risk or its charge. In the latter, policyholders collectively face risks and their material consequences. Risk-shifting mechanisms extract their efficiency by trading risks from risk-adverse agents or agents who have few resources for facing risks to less risk-adverse or more affluent agents. Risk-pooling mechanisms extract their efficiency from the Law of Large Numbers (cf. infra note 14), i.e. the increased statistical reliability implied by the pooling of risk profiles and resources.
For a transfer of costs to be completely fair, some conditions should be met regarding the bargaining conditions that led to the agreement. For instance, A should not be pressured by B to accept the deal. In addition, there are two issues here: one of consent, which is the most important, and another of compensation since some situations where a person suffers some additional costs might be fair everything considered because she is compensated despite not having agreed to it. It might be argued that, under certain circumstances, such an arrangement is fair. But, it is not discussed here.
Expected utility may be objective or subjective, depending on if it includes pure facts or one’s conviction. Some prefer the distinction between frequency and belief (Hacking 2001, pp. 127–139). In any case, according to the frequency-type (objective), probabilities are testable, impersonal and based on tendency, propensity and disposition. According to the belief-type, probabilities express a personal or interpersonal view based on confidence and evidence. Except when mentioning that probabilities should be understood from the individual perspective, we use expected utility in an objective fashion, following the frequency-type probabilities.
In other words, if insurance may be seen as redistributive at a given moment in time, it is not redistributive in terms of expected utility.
The Law of Large Numbers is the ‘mathematical premise stating that the greater the number of exposures, (1) the more accurate the prediction; (2) the less the deviation of the actual losses from the expected losses (X − x approaches zero); and (3) greater the credibility of the prediction (credibility approaches 1). This law forms the basis for the statistical expectation of loss upon which premium rates for insurance policies are calculated’ (Rubin 2008, pp. 272–273). In other words, ‘[a]s the number of trials increases, the accuracy probability approaches 1’. Then, ‘[r]elative frequencies tend to converge on probabilities’ (Hacking 2001, p. 197).
We are grateful to Søren Midtgaard for bringing this point to our attention.
Community rating takes place when no differentiation is made among contributors in regard to their risk exposure.
Moral hazard characterizes a ‘circumstance which increases the probability of loss because of an applicant’s personal habits or morals; for example, if an applicant is a known criminal’ (Rubin 2008, p. 322).
(As evoked below, this objection holds only insofar as risks are choice sensitive. Moreover, moral hazard takes place no matter if premiums are actuarially fair or not. It illustrates the fact that moral hazard is an argument against insurance in general, not against non-actuarial premiums in particular.
The view discussed here introduces individual responsibility in the calculation of the adequate premiums in the sense that fair premiums should reflect one’s level of responsibility in regard to the risks covered. It is, however, not the only use of the concept of responsibility. For instance, one may claim that rather than adjusting one’s premiums, individual responsibility should be used for drawing a line between insurable (responsibility-insensitive) risks and uninsurable (responsibility-sensitive) ones.
Case C-236/09, Association des Consommateurs Belges Test-Achats ASBL and Others v. Conseil des Ministres (2011).
‘ECJ gender ruling hits insurance costs’, The Guardian, March 1st 2011.
For a detailed discussion, refer to Heath (2007).
City of Los Angeles Department of Water and Power v. Manhart, 435 U. S. 702 (1978).
This touches upon the broader issue of the quality of the scientific standards used by risk management.
Economies of scale happen when adding more workers increases the global productivity of the unit of production (e.g. the factory) in a larger proportion than the sum of their individual productivities.
The question of the connection between health insurance and labour mobility revolves around the question of the portability of the rights to health insurance (e.g. Bailey and Chorniy 2013; Holtz-Eakin 1993; Madrian 2004). We are grateful to one anonymous referee for indicating such a connection to us.
A’s activities may enhance the situation of other agents, which is a Pareto improvement. However, it might be the case that, due to strict actuarial premiums, by undertaking his activity, and while improving the situation of everyone else, A is actually worsening his own situation. This advocates for accounting for spillover and external effects in any normative evaluation of insurance and the fairness of its conditions.
One may argue that actuarial premiums are not the problem here, but the fact that a given activity has a spillover effect from which the producer of the activity does not benefit. The distinction would be blatant in the obvious solution of re-internalizing the externalities by transferring additional resources collected by the distant beneficiaries of the positive externalities to their producer. Therefore, premiums could remain determined on a strict actuarial basis whereas beneficial activities could be properly incentivized. However, this kind of measure precisely shows that actuarial premiums are incomplete in the sense that they need to be corrected (or complemented) to be fully fair and efficient. In other terms, actuarial fairness does not encapsulate the complete story of the fairness and efficiency of insurance as a mechanism of social cooperation.
Ainslie, G. (2001). Breakdown of will. New York: Cambridge University Press.
Ariely, D. (2008). Predictably irrational: The hidden forces that shape our decisions. New York: HarperCollins.
Arrow, K. J. (1963). Uncertainty and the welfare economics of medical care. American Economic Review, 53(5), 941–973.
Baker, T. (1996). On the genealogy of moral hazard. Texas Law Review, 75(2), 237–292.
Baker, T. (2000). Insuring morality. Economy and Society, 29(4), 559–577.
Baker, T., & Simon, J. (Eds.). (2002). Embracing risk: The changing culture of insurance and responsibility. Chicago: Chicago University Press.
Bennett, C. (2004). Dictionary of insurance. Harlow: FT Prentice Hall.
Daniels, N. (1990). Insurability and the HIV epidemic: Ethical issues in underwriting. The Milbank Quarterly, 68(4), 497–525.
Daniels, N. (2004). The functions of insurance and the fairness of genetic underwriting. In M. A. Rothstein (Ed.), Genetics and life insurance: Medical underwriting and social policy (pp. 119–145). Cambridge, MA: MIT Press.
Dworkin, R. (2002). Sovereign virtue: The theory and practice of equality. Cambridge, MA: Harvard University Press.
Ewald, F. (1986). L’État providence. Paris: Grasset.
Goodin, R. E. (1988). Reasons for welfare: The political theory of the welfare state. Princeton, NJ: Princeton University Press.
Hacking, I. (2001). An introduction to probability and inductive logic. Cambridge, UK: Cambridge University Press.
Heath, J. (2006a). The benefits of cooperation. Philosophy and Public Affairs, 34(4), 313–351.
Heath, J. (2006b). On the scope of egalitarian justice. Les Ateliers de l’Éthique-The Ethics Forum, 1(1), 21–41.
Heath, J. (2007). Reasonable restrictions on underwriting. In P. Flanagan, P. Primeaux, & W. Ferguson (Eds.), Insurance ethics for a more ethical world (Research in ethical issues in organizations) (Vol. 7, pp. 127–159). Bingley: Emerald Group.
Holtz-Eakin, D. (1993). Is health insurance crippling the labor market? Public policy brief 10. New York: The Jerome Levy Economics Institute of Bard College.
Hoy, M., & Ruse, M. (2005). Regulating genetic information in insurance markets. Risk Management and Insurance Review, 8(2), 211–237.
The Guardian. (2011). ECJ gender ruling hits insurance costs. March 1.
Kahneman, D. (2011). Thinking, fast and slow. New York: Penguin.
Knight, F. (1921). Risk, uncertainty, and profit. Boston: Houghton-Mifflin.
Lane, W. R. (1997). Discussion of T.A. Moultrie and R.G. Thomas’s “The right to underwrite? An actuarial perspective with a difference”. Journal of Actuarial Practice, 5(1), 141–146.
Leigh, S. (1998). The freedom to underwrite. In T. Sorell (Ed.), Health care, ethics and insurance (pp. 11–53). London: Routledge.
Lethonen, T. K., & Liukko, J. (2011). The forms and limits of insurance solidarity. Journal of Business Ethics, 103(1), 33–44.
Madrian, B. (2004). Health insurance portability, labor supply and job mobility. In Inter-American conference on social security CISS/WP/04012.
Meyer, R. B. (2004). The Insurer perspective. In M. A. Rothstein (Ed.), Genetics and life insurance: Medical underwriting and social policy (pp. 27–47). Cambridge, MA: MIT Press.
Moss, D. A. (2002). When all else fails: Government as the ultimate risk manager. Cambridge, MA: Harvard University Press.
Moultrie, T. A., & Guy, T. (1997). The right to underwrite? An actuarial perspective with a difference. Journal of Actuarial Practice, 5(1), 125–138.
Papandreou, A. A. (1994). Externality and institutions. New York: Oxford University Press.
Radetzki, M., Radetzki, M., & Juth, N. (2008). Genes and insurance: Ethical, legal and economic issues. Cambridge, UK: Cambridge University Press.
Rawls, J. (1958). Justice as fairness. The Philosophical Review, 67(2), 164–194.
Rubin, H. W. (2008). Dictionary of insurance terms. New York: Barron’s, Hauppage.
Rubinstein, A. (1998). Modeling bounded rationality. Cambridge, MA: MIT Press.
Simon, H. (1957). Models of man: Social and rational. New York: Wiley.
Stanovich, K. E. (2009). Decision making and rationality in the modern world. New York: Oxford University Press.
Stone, D. (1999–2000). Beyond moral hazard: Insurance as moral opportunity. Connecticut Insurance Law Journal, 6(1), 11–46.
Taleb, N. N. (2008). The black swan: The impact of the highly improbable. London: Penguin.
Tversky, A., & Kahneman, D. (1974). Judgments under uncertainty: Heuristics and biases. Science, 185(4157), 1124–1131.
This research received the support of the Danish Council for Independent Research, project Public Insurance, Equality, and Efficiency (10-080448). This paper benefited from the comments of two anonymous referees of the Journal of Business Ethics, Axel Gosseries, Nils Holtug, Kasper Lippert-Rasmussen, Martin Marchman, Søren Midtgaard, Morten Nielsen as well as comments received at several occasions such as the seminar of the Department of Political Science at Aarhus University, the CESEM seminar, the Workshop on Insurance and Discrimination (University of Copenhagen) and the 7th International Conference on Applied Ethics in Hokkaïdo.
About this article
Cite this article
Landes, X. How Fair Is Actuarial Fairness?. J Bus Ethics 128, 519–533 (2015). https://doi.org/10.1007/s10551-014-2120-0
- Expected utility