Alongside private insurers sit carriers of another species, public export credit agencies and multilateral bodies attached to development banks, and what they bring must be understood before they are compared on price. Their peculiarity lies neither in their capacity nor in their wordings, often close to those of the private market. It lies in their shareholders: the state hosting the investment is often a member of the institution insuring it, or borrows from its parent. The guarantee then stops being a mere transfer of risk and becomes a political fact, and it is that second nature which is paid for and which makes the choice interesting.
The most useful effect goes by a name the subject uses without always explaining it, the deterrent effect. Taking an asset insured by a body of which one is a member, or whose parent finances one's roads and hospitals, exposes a state to a consequence belonging not to litigation but to credit: the measure becomes a subject in its relations with lenders it does not wish to antagonise. A private policy has nothing of the sort to offer. It indemnifies better or worse, faster or slower, but it does not change the probability of loss. A multilateral guarantee changes it, marginally, and that is what is bought.
What this costs must be measured at once, because the price is not in the premium and that is what misleads. Processing is long, several months where a private insurer commits in weeks, and a transaction timetable is constrained accordingly. Eligibility conditions are substantial: environmental and social criteria, consultation of affected populations, sometimes the host country's own agreement, which means telling it that one is insuring against it. Projects already completed are frequently ineligible, the institution's logic being to encourage new investment rather than to cover an existing one. And monitoring is continuous, with reporting obligations lasting as long as the guarantee.
A second difference deserves stating plainly because it inverts an intuition. Against a private insurer, a dispute is settled by arbitration or before a judge, with known delays and costs. Against a multilateral body, part of the resolution runs through internal and diplomatic mechanisms, sometimes more effective and always less predictable, where the insured's position depends on considerations beyond it. One therefore exchanges an adversarial procedure for influence, which is excellent while the influence runs the right way and uncomfortable when it runs elsewhere.
Then comes the practical question, that of two offers at the same price, and the temptation to handle it by comparing wordings must be resisted. Both policies may cover the same perils with similar definitions; what separates them lies elsewhere. Available duration first: public carriers routinely accept commitments of fifteen or twenty years the private market does not grant, which matters for an asset depreciated over that period. Stability of renewal next, a private insurer being able to withdraw from a country at the following expiry when a development institution stays. And the link with financing, several lenders lowering their capital requirement in the presence of a guarantee of this family, which moves the comparison outside the insurance budget.
A third actor is often met and must not be confused with the first two, the export credit agency of the investor's own country. Its mandate is not to protect the investment but to support national exports, which makes it available when the project buys equipment from its country and unavailable otherwise. Its guarantee therefore follows the geography of the supply chain rather than that of the risk, and an investor unaware of this builds its cover plan on an availability depending on a purchasing decision it has not yet made.
It remains to say what such a carrier's presence changes in a claim, because that is where expectations are disappointed. A multilateral institution investigates rigorously and slowly, applies its own rules of evidence, and is attentive to not creating a precedent towards its other member countries. An insured hoping for rapid payment because its insurer is powerful discovers the opposite: the power serves to prevent the loss, not to settle it quickly. Both effects are real and they do not operate at the same moment, one before, the other after, and it is the confusion between them that produces disappointment.
The honest comparison is therefore made on a single axis and it is temporal. If the exposure is short and the need for rapid payment real, the private market serves better. If the asset is immobilised for fifteen years in a country where what one wants above all is that nothing happen, the public carrier serves better, provided one accepts long processing and real conditionality. Two offers at the same price are not two comparable offers, and the useful professional is the one who moves the discussion from the amount to the duration and to what each carrier can do before the loss.
A European manufacturer is building a 320 million euro fertiliser plant in an East African country, financed 70 per cent by a pool of international banks, depreciated over eighteen years. Two political risk offers arrive for a 200 million guaranteed amount at an identical annual cost. The first is private: a three year renewable commitment, underwriting closed in five weeks, arbitration in London, contractual settlement within ninety days of a completed claim file. The second comes from a multilateral body of which the host country is a member: a firm fifteen year commitment, processing estimated at five months with an environmental and social impact study and formal host country consent, no stipulated settlement period. The bank pool indicates it would reduce its margin by 35 basis points in the presence of the second. What should be advised?
Comparing the two offers by their identical annual cost is the error to name at the outset, because they are not two products of the same kind and price is not what separates them. The decisive fact is duration: the asset is immobilised for eighteen years and the first offer commits for three. The insured would therefore carry renewal risk for fifteen years, that is precisely the risk that materialises as a country deteriorates, a private insurer being free to withdraw at the next expiry and generally doing so at the moment cover becomes useful. The second fact is the deterrent effect, which is not a sales argument: the host country is a member of the institution, and a takeover of the plant would become a subject in its relations with its lenders. On an immobile asset that can be neither moved nor sold, reducing the probability of loss is worth more than accelerating its settlement. The third fact falls outside the insurance budget and is the heaviest: 35 basis points on 224 million of debt over eighteen years represents an amount that dwarfs the gap between the two premiums, and it appears on no cover comparison table. Two reservations must be put frankly to the client rather than discovered later. Five months of processing and formal host country consent are incompatible with a tight construction timetable, and one must check whether the banks' commitment holds that long; that is the point on which the transaction can genuinely fail. And the absence of a stipulated settlement period is no oversight: this family of carriers investigates at length and is mindful of its other member countries, so the insured must know it is buying prevention and duration, not speed of payment. If the timetable does not hold, the construction that answers the file is not to take the short offer but to cover the construction period with the private policy and move to the long guarantee at commissioning, which means filing the application now.
- 01A multilateral guarantee is distinguished by its shareholders and not by its wording: the insured state is often a member of the institution insuring it.
- 02A private policy does not change the probability of loss, it organises its settlement; a multilateral guarantee changes it marginally, and that is what is bought.
- 03The price is not in the premium: months of processing, environmental and social conditionality, host country consent, frequent ineligibility of completed projects.
- 04Available duration and stability of renewal separate the offers far more than price, and the effect on the cost of debt falls outside the insurance budget.
- 05A carrier's power serves to prevent the loss, not to settle it quickly: the two effects do not operate at the same moment.