How does one establish the cost of risk? Two theories apply. Under the first one, of a largely psychological nature, the cost of risk is measured by the aversion to risk of the individual who wishes to avoid it. In this case, quantification is not required. The cost of risk is the total premium agreed upon between the individual who wishes to transfer it and the one who agrees to assume it. The following anecdote illustrates this point. In an effort to promote his whisky, the president of the Cutty Sark distillery had offered a large monetary sum to the first person to see the Loch Ness Monster. It may be that one night he had a nightmare or suffered remorse, but he would have to keep his foolish promise none the less, which would undoubtedly endanger his company’s financial well-being. He therefore hastened to find an insurer, someone to whom he could transfer the risk he had taken so brashly. And he found one, Lloyd’s of London, with which he agreed on a premium. The coverage of this unsurpassable, potential, and exceptional risk involved a cost — a risk transfer. There was insurance without any need to quantify. The concept of insurance risk is not so much tied to the notion of danger as it is to that of expectation or fear. Risk is the measure of an expectation — a mathematical expectation that, according to Pascal, is the product of the probability of the event multiplied by its value. A moral expectation…which relates to what I am willing to and what I ought to pay, in terms of mathematical expectation in order to avoid risk. The “additional cost” would measure exactly what the value of the risk is for me.

In the words of Michel Albert who initially made the distinction, this purely contractual vision of insurance corresponds to a more “maritime” than “riparian” view of insurance. In continental Europe, France in particular, one views insurance as tied to the concept of mutuality, or as a function of statistics and probability. This view stems from the fact that, in France, insurance had to set itself apart from gambling and betting to be recognised. It follows that risk reflects the probability that an event will take place in a given population. Consider the population of French drivers, the road conditions, and the number of cars on the road. There are currently about 8,000 annual road deaths. In the words of the well-known sociologist Adolphe Quetelet, these deaths represent the “budget” that the French population commits to road travel each year. One can anticipate that, all things being equal, this figure will remain constant from year to year, within a few standard deviations. The average premium per inhabitant or per driver can then be established. At the same time, it is possible to assess the chance that any one individual has of being among the 8,000 victims. This chance is a function of a car’s power, a driver’s experience, the places where he drives and so forth, and these criteria are used to calculate an individual’s insurance premium. Indeed, not everyone represents the same risk to the group. Some individuals drive less carefully than others. This is another way of establishing the cost of risk; it corresponds to the greater or lesser probability of incurring an accident in relation to the average. This cost is called the equitable cost.

Under this second analysis (more sociological than psychological) the group outweighs the individual. Risk influences the group. It affects it with depressing regularity and affects every individual as members of the group. Risk has a defining and unifying effect on the group and gives it a personality and an identity. This is the paradox of freedom: each member can feel as free as he will; through his actions he will contribute in one way or another to reproducing the common statistics. Such a group, identified on the basis of its risk, is what is called a mutuality.

By assessing the cost of risk so that it can become the subject of a contract, insurance gives existence to non-existence, immediacy to the non-current, and form to potential. It inverts the course of time. The feared event exists whereas it has not taken place and may in fact never take place. Risk exists, materialised through a contract, in the form of an infinitesimal fraction of what it will become. That is, it exists in such a manner that its presence, while thoroughly real, remains almost imperceptible. This is what makes insurance doubly effective. The fragmentation of that risk renders feared events no longer feared. Insurance eradicates and decimates adversity. It makes these fears more bearable. Sometimes too much so.

But the merit of insurance extends beyond what would otherwise closely amount to a public-aid dynamic. It transforms the perception of events in such a way so that not to carry insurance becomes a fault. This point was well illustrated last century by Edmond About in his booklet insurance [L’Assurance (1865)]:

Insurance suggests a social experience of risk to both liberalism and democracy. To the former because liberalism is a political philosophy that advocates risk management as a principle of government. Individuals must face risk in order to become truly aware of their real identity; finding resources in themselves in the form of foresight and, in others, in the form of voluntary association. But insurance, at least in its practical form, is also the product of democracy to which it gives its image of solidarity. Thus, for two centuries, insurance has continuously prompted us to be aware of ourselves both individually and collectively. As the Baron de Beauverger said in 1868 during a parliamentary debate on industrial accidents: “As a system, our society is nothing but an all encompassing insurance, insurance against weaknesses, insurance against misfortune, insurance against ignorance. Look at institutions through that prism and you will see that they all seek the same goal, a noble and generous goal.”

This analysis of risk was seriously undermined when an attempt was made to make insurance resolve social problems linked to the development of an industrial society. With the rise of social insurance, insuring oneself becomes mandatory…It is a matter of seeing to it that individuals are protected against certain risks of a welfare nature. The concept is twofold: regardless of his income, every individual must be protected from these risks since such risks are not equally allocated throughout society…. If there are social risks, it is because society itself generates them as it develops without regard to a fair allocation. Compensating for inequalities in the face of risks, establishing equalities in the face of opportunities…. Social insurance makes coverage against social risks no longer an act, but an entitlement. It is first a salaried worker’s entitlement, and the institution of social security makes it a citizen’s entitlement. The actualisation of the potential, of which the economic translation is capitalisation, gives way to a sort of generalised assistance based on allocation.

This rationale regarding insurance became extremely widespread. Today, welfare outlays in the nation of France exceed the budget of the state of France. And the gap continues to widen in accordance with a rationale under which the nature of covered risks (health and retirement) hardly suggests a shift. Yet, the limit beyond which welfare outlays seem destined to ruin their own sources of supply has been reached: social costs are exhausting the economy. The economic rationale couples with a moral rationale: instead of creating an incentive to take risks, health and welfare services engender phenomena of “demoralization” that are counterproductive. Insurance no longer operates as an incentive to take risks, indeed, it directs us to never have to take any.

We have considered some of the issues regarding the necessary limit to risk-taking and the danger of excessive risk-taking. We now find ourselves facing an inverse situation: what we may fear is not excessive risk-taking, but, on the contrary, the absence thereof. Once again, there arises an issue of balance.

Today, we are witnessing a remarkable shift in this pattern. The issue is no longer so much that of multiplying risk liabilities and structuring through insurance the solvency of those liable as it is to prevent certain risks from being taken. Not only is prevention outpacing indemnification, but efforts are also made to avert risks that have not yet been recognised. Precaution governs. Several factors account for these recent developments. First, damages no longer pertain to individual accidents as much as they pertain to catastrophes. The amounts currently involved exceed the limits of what can be insured as well as indemnified. Secondly, the cost of liability is also being reevaluated. The First World War provides a good scale by which to weigh this new method of measuring risk. During the war, a general could send 300,000 men every two weeks off to be killed, as was the case during the Chemin des Dames battle. Today, only “zero-risk” wars can be waged. A peculiar transmutation of values, indeed! Under the traditional cost-benefit analysis, it was enough that advantages outweighed risks to feel justified in taking risks and thus in accepting a portion of loss. Today, risk tends to be measured on the basis of the loss portion: what justifies the sacrifice? Do not those unfortunate enough to make up the loss portion count as much as others? Such is the method of valuation underlying the zero-risk problematic.

The rationale of precaution does not advocate, as it has been said, a change of focus from risk to fault. It results from a twofold reassessment of risk. First, it results from the technological powers that are now at work and over which we know we fail to exercise full control. We are witnessing an excess of might over power that we do not really know how to express from a legal standpoint. It can be characterised as a developmental risk, principle of precaution depending on whether we look at it from a legal or political standpoint. Secondly, the rationale of precaution results from a sort of backlash of victims who no longer accept the cynicism arising from the traditional formula for the acceptance of liability. What they challenge is not so much the amount of damages as the imbalance of power linked to technological risks. And, today, as a result of this backlash, liability tends to be assessed on the basis of what was hitherto considered negligible.

Here again is another case of how the experience of risk has its limits. Given the inordinate power now in existence (which can be measured by the fact that we cannot measure its effects), the relations of asymmetry and dependence (with the sense of a loss of autonomy that ensues), and the magnitude of the risks created, should it be appropriate, short of putting an end to the process, at least to take a pause that would allow us to regain control over the changes governing us? Time for precaution is time for moratoria.

The perception of risk constitutes a defining experience for contemporary society: how far is too far? While valuing risk, adventure and entrepreneurship, contemporary society seeks to keep it within measure. Thus, while risk stands as a principle of valuation, motivation, and action, it also constitutes a principle of limitation, restriction, and prohibition. When overvalued or undervalued, risk quickly turns human experiences into inhuman ones. There is therefore no need to set a morality of risk against a morality of protection. Indeed, the morality of risk is inextricably a morality of protection. Risk and safety are not opposite concepts independent from each other, Risk both affirms and negates. It arises from the need to surpass oneself, from the necessity to tran...