You cannot insure a belief

The reflex is understandable. Anything valuable that can be destroyed calls for insurance, and reputation ticks both boxes with a new brutality. An executive sees their name as capital, notes that a rumor can eat into it faster than a fire consumes a warehouse, and logically seeks to transfer that risk as they transfer the others. Brokers offer policies that carry the word reputation in their title, insurers answer, a market exists. And yet, looked at closely, this market never quite insures what it claims to insure, and the gap will not close as the market matures, because it comes from the nature of what is being insured.

Here is the thesis. Reputation is the one asset whose entire value lives in the belief of others, and that is precisely why it resists insurance at the root. Insurance needs an external, objective loss, something that exists independently of what anyone thinks. Reputation has no existence outside opinion, it is what people think of you, so the loss happens in the minds of customers and journalists, and nowhere else. You cannot indemnify a belief. What the market sells under the name of reputation insurance never covers the reputation, only its measurable shadows.

Why one insures a building and not an opinion

The most illuminating comparison is the building. One can insure a building because it exists apart from any gaze, its walls stand whether one thinks of them or not, and its damage can be observed, measured, priced. Reputation is the exact opposite, it has no walls, it exists only in other people's heads, and its value consists solely of the trust placed in it. To destroy a building is to alter the physical world. To destroy a reputation is to alter a shared belief, and a belief is not repaired the way one rebuilds a wall, it does not come back because a check has been signed.

From this difference everything else follows. The loss of reputation has no clear outline, one cannot say where it begins or ends, nor exactly how much it cost, because it merges with the shifting mood of a public. It is also reflexive, the very announcement that a reputation has been harmed can harm it further, so that measuring the damage sometimes adds to the damage. An asset whose value cannot be fixed before the loss, whose loss cannot be fixed after, whose bounds cannot be fixed during, is an asset insurance cannot grasp, because insurance lives on numbers and reputation lives on judgments.

Beneath this impossibility there is even a problem of ownership. An asset can be insured only if one holds title to it, yet no one owns their reputation. It barely shows on the balance sheet, except buried in the goodwill of an acquisition, and it actually resides in the minds of millions of people who owe you nothing, customers, employees, journalists, strangers, and who can alter it without your consent. You are the subject of your reputation, you are not its owner, and one does not insure an asset whose title belongs to the crowd. It is a value you count among your assets but do not hold, lodged in others, and this split between the one who suffers the loss and those who decide it has no equivalent in the classic lines.

What the market really covers

One must then look at what these contracts actually cover, because they exist and they pay. What they indemnify is never reputation itself, it is the measurable trail it leaves as it falls, the revenue lost after a scandal, the cost of a crisis unit, the fees of the communications agencies mobilized to put out the fire, sometimes a share-price fall tied to a precise event. In other words, the market has silently given up on insuring the asset, and insures instead its accounting consequences, which is an elegant way of admitting that the object escapes it.

This admission can be read in the very structure of the covers. To pay anything, the insurer needs an objective trigger, a threshold of media coverage crossed, a drop in sentiment measured by an index, an attributable fall in revenue. It sets the contract on a shadow because it cannot set it on the thing. This is the logic of parametric cover applied to an intangible asset, and it inherits all its known flaws, a gaping basis risk between the chosen index and the harm truly suffered, a payout that may fall when reputation has not moved, or stay silent when it has collapsed without the chosen indicator reacting. One measures the smoke and hopes it reveals the fire.

One comparison completes the picture of the boundary. One easily insures the legal costs of a defamation suit, and that insurance works very well, because it bears not on reputation but on an objective fact, a precise statement, false, published, whose wrongful character a court will assess and whose damages it will fix. There is a judge, a standard, a figure. Defamation is a bounded, adjudicable, priceable wrong, everything reputation is not. To insure the lawsuit is to insure a thing in the world, to insure reputation would be to insure the mood of the world, and that is exactly where the line runs between the insurable and the uninsurable, between the wrong one proves and the opinion one suffers.

The risk you inflict on yourself

Reputation resists insurance for a deeper reason still, it is endogenous, it depends on the very party that would protect itself. A building burns by accident, a reputation is most often lost by conduct, it is earned and forfeited, it is the accumulated reflection of what one has done. To insure a reputation would therefore be to insure someone against their own behavior, which collides head-on with the principle of moral hazard. Worse, the cover could aggravate the risk it claims to carry, for a firm that knows itself shielded from the fallout of its faults may watch itself a little less, and insurance would then become a reward for reputational recklessness.

To this endogeneity is added the reflexivity of the damage. Reputation is lost through the triggering event, then through the reaction to the event, and that reaction the insured partly controls. A well-handled crisis limits the harm, a badly handled one multiplies it, so the size of the loss depends on the competence of the victim as much as on the gravity of the facts. The insurer would find itself underwriting not an external peril, but the character and skill of its insured, and no actuary prices a chief executive's composure in front of the cameras. There is a circle here that no contract closes, the loss is partly made by the one it strikes, and the very triggering of the cover, by signaling the scale of the harm, can deepen the wound it was meant to heal.

The fire that outruns the truth

All this existed before, but a recent acceleration has laid the problem bare. Social media has made reputation both more valuable and infinitely more fragile, by crushing the scale of time. What was built over decades can be undone in hours, and the pace of destruction now bears no relation to the pace of construction. Insurance, which assumes losses slow enough to be observed, assessed, reserved, finds itself facing a peril that has consumed most of its effects before a file is even opened.

More troubling still, reputational damage has detached itself from truth. A false but viral accusation harms as much as a well-founded one, sometimes more, because the public reacts not to facts but to what it perceives of them, and the correction always arrives after opinion has formed. Reputation has thus become a risk that is fast, correlated, indifferent to truth, three properties that are so many poisons for insurability. Fast, it defeats the timing of assessment. Correlated, since one wave of outrage can strike a whole sector at once, it brings back the specter of the simultaneous loss. Indifferent to truth, it makes the very notion of a triggering fact uncertain, since the trigger may be a lie no court has yet refuted.

The score the crowd gives

Reputation is, at bottom, a behavioral score of a particular kind, except that the crowd does the scoring, and not the insurer. Where behavioral pricing entrusts a sensor with translating your conduct into a premium, reputation entrusts opinion with translating your conduct into value, and the verdict of opinion is faster, more arbitrary, less revisable than any insurer's algorithm. In the behavioral case, we measured the danger of a continuous judgment delivered by an opaque box. Reputation is the same continuous judgment, delivered by an even more opaque box, the collective mind, which no one can audit and whose formula no one knows.

This judgment can also strike a whole sector at once. When a scandal breaks at one player, suspicion spreads to its neighbors, one greenwashing incident stains the entire industry, a fraud found at one casts doubt on all the others. Reputation thus brings back the specter of the simultaneous risk seen with the pandemic, a common cause, here outrage, that at a stroke correlates losses believed independent. An insurer that thought it had diversified by covering the reputation of twenty firms in the same sector would discover, on the day of a sector-wide scandal, that it held only a single policy. The shadow that reputation insurance claims to cover is therefore not only blurry, it is also contagious.

The indices on which these contracts rest have, finally, the flaw of being manipulable. Media volume can be inflated, sentiment manufactured, a coordinated campaign can push past a trigger threshold without any real harm having occurred, or conversely drown a genuine scandal under favorable noise. The trigger of a reputation cover can thus be produced on purpose, which opens the door to a new kind of fraud, no longer exaggerating a loss that occurred, but fabricating the signal that commands the payout. One does not build sound insurance on a measure the insured, or their adversary, can stage.

Managing what one cannot transfer

From this uninsurability follows an instructive shift. Since reputation risk cannot be transferred, one returns to reducing it, and the market that thrives around reputation sells prevention, monitoring of weak signals, preparation of crisis units, media training of executives, mapping of vulnerabilities. Where classic insurance promises to repair after the fact, reputation demands work upstream, because after the fact it is already too late, the belief has tipped and no check brings it back. We meet again, inverted, a lesson encountered earlier, when a risk escapes transfer, all the value takes refuge in prevention and in the speed of response.

There is a final irony to ponder. The insurer is itself one of the firms whose reputation is the most vital capital, for it sells a promise, and a promise is worth only the trust granted to the one who carries it. An insurer whose reputation wavers, who is suspected of not paying, loses the essence of its trade long before it loses a lawsuit. So the one asked to insure the reputation of others itself owes its existence to an asset it could not insure for itself. Reputation is the ground on which all insurance is built, and one does not insure against the collapse of the ground that holds you up.

An asset one is not allowed to save

There remains the question that gives this issue its point. What if reputation ought to remain uninsurable, not through a failure of the market, but through a social necessity? Reputation serves not only the one who holds it, it serves everyone else, for it is one of the oldest mechanisms of discipline a society has without resorting to law. We keep our commitments in part because betraying them costs us our name, we tend to our conduct because the gaze of others carries a price. A reputation disciplines behavior only as long as it is genuinely at stake, as long as it can be lost.

To fully insure reputation would therefore be to extinguish what makes it useful. A firm perfectly covered against disgrace would have nothing left to fear from disgrace, and the name would cease to guarantee the conduct. Here we meet again, transposed, the paradox found in behavioral pricing, a risk one cancels ceases to discipline the one who bore it, and perfect protection destroys the social function of the peril. Reputation belongs to that rare family of risks a society has an interest in leaving in place, because it is their very threat that produces order. Here uninsurability marks a boundary, and the market is right not to cross it.

This is why reputation will remain, for the most part, beyond the reach of insurance, and why that is probably good news. One can insure the cost of answering the crowd's verdict, one cannot insure against the verdict itself, and it would be dangerous to be able to. The name of a company is the accumulated judgment of all who deal with it, and that judgment cannot be bought back, only earned, lost and earned again. A fully insured reputation would no longer be a reputation, it would be a line on the balance sheet, and a world that could insure against shame would have lost the cheapest form of order it possesses. Reputation thus remains the last asset money does not buy back, and this resistance, far from being a weakness of the market, states its function, there are things a society insists on leaving exposed so that they bind. One insures what can be replaced, one does not replace a name, one holds it or one loses it, and no policy has ever changed that.