It seems appropriate to begin this reflection by answering a basic but fundamental question: what exactly is meant by smart contract in the world of law? Law no. 12/2019 introduced a legal definition of smart contract, i.e. “a computer program that operates on technologies based on distributed registers and whose execution automatically binds two or more parties on the basis of predefined effects by the same”. In essence, we can think of smart contracts as the transcription of (contractual) clauses into algorithms or codes that automatically verify the fulfilment of certain conditions (control of basic contract data) and automatically perform specific actions when the conditions determined between the parties are reached and verified.
Smart forms of contract had already been in place since the mid ’70s when, for example, the licensing of some software was in fact handled by a digital key that allowed the software to run if the customer had paid for the license and had it cease to run on the expiry date of the contract. What makes this method of consensus building particularly interesting today is the fact that, through blockchain technology, guarantees of reliability, transparency, immutability and therefore certainty are given to the contract. As has been correctly observed, the smart contract needs legal support for its drafting, but does not need it for its verification and activation, and is in fact a program that processes in a deterministic way (with identical results under identical conditions) the information that is collected. Therefore, if on the one hand this represents a certainty and security as it guarantees the parties an absolute “certainty of objective judgement” excluding any form of interpretation, on the other hand it shifts on the code and programming, responsibility and power to decide (and therefore in essence certain issues related to the will of the parties).
Contractors are responsible for defining terms and conditions and terms and conditions and rules of control and action, but once their contract has become a code and therefore smart contract and the contractors accept it, the effects and its execution no longer depend on their will.
Indice degli argomenti
Practical applications: from insurance contract to smart insurance contract?
The insurance industry has already been looking with interest at smart contracts for some time now: the question therefore arises as to whether it is possible to “build” an insurance contract (a “product”) that, once concluded, can be executed on its own, without further action, intervention or infrastructure.
Initiatives in this direction are indeed beginning to multiply. An example is given to us by Etherisk, a policy that operates on the blockchain Ethereum platform and covers the risk of flight delays, based on mere data provided by the airports of departure and destination. The smart policy, therefore, verify that the “instructions” have been fulfilled (flight XY to Paris was supposed to land at 6pm and instead landed at 9pm) will automatically settle the compensation. A policy to cover the ruined stay in case of bad weather (automatic reimbursement in case of bad weather, based on the third source of weather forecasts) has also been tested.
The subject, of course, arouses a lot of interest but also raises several questions, due to some peculiarities that arise from the comparison with the civil law governing the insurance contract. Without claiming to be exhaustive, we have therefore turned our attention to some fundamental issues.
The principle of indemnity and the verification of damages
As is well known, non-life insurance contracts are based on the principle of indemnification, whereby both the insured party was damaged by the risk event covered and the insurer, to the extent that it is obligated, settles the indemnity.
The contract, therefore, needs two basic elements to be executed: (i) the materialisation of the risk deducted and (ii) proof (and quantification) of the loss or damage for which the policyholder is liable (as compensation may not exceed this amount).
Is it possible to translate these operations into codes/algorithms, with a gain in terms of speed, transparency and certainty of compensation?
In other words, can compensation be paid on the basis of a lump sum quantification identified by the parties, on the basis of an indirect investigation linked more to the cause (e.g. the damaging event) than to the effect (e.g. the resulting damage)?
The margins, in our opinion, are: in particular in cases where the risk deduced can be objectively ascertained (e.g. the delay of an aircraft or precipitation in a certain area) and the damage is immanent in the cause itself (e.g. in the case of delay), or otherwise susceptible to presumption (in the case of a ruined holiday). The problem will, if anything, be linked to the reliability of the sources to which we are conventionally entrusted to ascertain the causal fact (the so-called “oracles”, in the blockchain language) and to the binding force of the clauses that make this clause, compared to the consumer discipline.
While the oracle can be a valid solution in general, it is not always a guarantee of compliance with the indemnity principle for all types of insurance coverage, particularly when the damage refers more precisely to the objective sphere of damage to property: think, for example, of a theft/fire policy, where it is necessary for the insured to document the lost property and its specific value (due to the concrete level of wear and tear). It is then easy to predict that a possible smart contract will be more easily applicable to the coverage of assets with market value subject to quota/list, on deposit with third parties (e.g. banks or authorized traders) with a guarantee of reliability. Unless, of course, the insured person is asked to confirm that there are no other guarantees for the same insured property (or, who knows, tomorrow find confirmation directly from the insurers, as is already the case in part today for the CAR).
Declarations of the insured
Another challenge for the smart contract insurance industry is the declaration of the policyholder during the underwriting phase (i.e. when the premium is calculated on the basis of the conditions represented by the policyholder).
As is well known, inaccurate and reticent statements by the insured, if made with intent or gross negligence, entitle the insurer to deny compensation, in particular whenever the (omitted) information has influenced the determination of risk. In the field of health policies, for example, the medical history questionnaire is the main tool through which information is collected so that the insurer can give its consent and determine the premium, to the extent that case law has held that it can assume that what is not mentioned in it is not in fact relevant for the insurer (who, as a professional, is required to guide the other party in identifying the information to be communicated).
Hence two fundamental fears: that the underwriting processes of smart contracts may take on the same probative value as anamnestic questionnaires (putting insurers in a position of extreme negotiating weakness); and that such rigidity ends up limiting the very adoption of smart contracts, whenever insurers do not feel comfortable closing the pre-contractual information they need to accept the insurance “bet” in a closed and predetermined set.
In conclusion
As can be seen, the challenges for the insurance industry (and legal practitioners) are still multiple. However, it is considered positive and indeed very appreciable that the market is studying the phenomenon in depth and experimenting with solutions to be proposed, because where this leads to speed and transparency in compensation, the first to benefit are the policyholders (first and foremost in terms of premiums required).
In order for smart policies to be implemented with the necessary guarantees of legal discipline, however, it is necessary to continue research: it is clear that there are still many legal aspects to be explored and the practices of the (secular) insurance market to be carefully assessed; this is certainly a great stimulus for legal scholars, for computer scientists and for those who study social phenomena.
A little anecdote
In 2011 a novel entitled “The index of fear”, written by Robert Harris, was published. Although the author is not here at his highest level, the book is interesting because it tells, in an imaginative and even improbable way, the story of a mathematical professor, owner and deus ex machina of a company that manages hedge funds. The protagonist in particular has created an absolutely innovative and top secret program that is based, to decide where to move the investments, on vix, (the so-called fear index), which measures the expectations of market volatility. The problem of the brilliant professor arises when the program, totally autonomous and self executing, starts to process wrong information and on such incorrect premises adapts the investment policies…
Verifying that the premises and initial assumptions on which the automatism of the program is based are correct and truthful will always require “human” control. Just as the “human” assessment can make a decision based on equity, which transcends binary parameters based on “if-then” principle.
All rights reserved
