Smart Contracts and Complex Transactions. A Good Fit?

Table of Content

A. INTRODUCTION

B. THE FUNDAMENTALS OF THE BLOCKCHAIN AND SMART CONTRACTS
I. The Blockchain
1. General Description 2. Potential Threats
II. Smart Contracts
1. General Description
1.1 Illustration of the Code
1.2 Visualization of a Smart Contract
2. Platforms
2.1 Economic and Legal Perspective of a Blockchain Platform
2.2 Bitcoin
2.3 Ethereum
2.3.1 General Information
2.3.2 The DAO Incident

C. Potential Fields of Application
I. Financial Services
II. Insurance
III. Healthcare
IV. Conclusion

D. Compatibility of Smart Contracts with Complex Contracts
I. Definition of a Complex Contract
II. Legal Perception of Smart Contracts
II. Fees and Capacities of the Blockchain
III. The Language of the Code
1. Semantic challenges
2. Ricardian contracts
3. Smart Contract Templates and Domain Specific Languages
4. Declarative Smart Contracts
5. Legal Enforceability of Programming Language
IV. Incomplete Contracting and Behavioural Challenges
1. Inclusion of Non-enforceable Terms in Contracts
2. Inclusion of Vague Terms in Contracts
3. Intentional Non-enforcement
4. Efficient Breach
5. Reduced Legal Uncertainty
6. Modification of the Contract and the Hold-up Problem
7. Conclusion
V. Privacy and Information Disclosure
1. General concern
2. Theoretical Examples of Information Disclosure Challenges
3. Conclusion
VI. Contracts Relying on External Services
1. Oracles
1.1 Oracles and Public Information
1.2 Oracles and Non-public Information
1.3 The Machine Learning Solution
1.4 Conclusion
VII. Termination of a Smart Contract

E. CONCLUSION

A. Introduction

After being evaluated as simply a matter for ‘techies’ for quite some time, many experts currently share the opinion that the blockchain is bound to become a disruptive technology for the transfer of digital assets and data storage. According to Tapscott, it even is “the technology most likely to change the next decade of business”.¹It could potentially replace centralized institutions and intermediaries with a decentralized computer architecture and could be a massive improvement of the way people are contracting and enforcing their claims, which would lead to a higher socioeconomic equality. However, one of the most interesting use cases of the blockchain technology is the implementation of smart contracts on it. Contracts like this are supposed to reduce transaction costs and make contracting more efficient. So far, smart contracts mostly conduct simple transactions that typically transfer ownership rights in response to conditional statements to do B if A occurs. Relatively simple transfers like this are unlikely to raise serious legal issues. But with time and the rapidly growing complexity of blockchain platforms, the complexity of smart contracts - as well as the associated possibilities and risks - might increase as well. In that respect, it is questionable whether smart contracts might be a suitable solution for convoluted transactions like multi-party transactions or transaction where the respective execution relies on external input. To answer this question, this thesis firstly tries to explain the core functioning of blockchain and smart contracts and subsequently will assess whether smart contracts are currently already ready to carry out complex transactions and if not, review and develop solutions for the prevailing challenges. In this context, it will analyze the possibilities as well as the legal and technical boundaries of smart contracts.

B. The Fundamentals of the Blockchain and Smart Contracts

In this section, this thesis explains the main idea behind the blockchain technology and how it works from a technological perspective. Subsequently, I will describe the arguably most interesting use case of it, smart contracts.

I. The Blockchain

1. General Description

Satoshi Nakamoto introduced the blockchain in 2008 and described it as a distributed and decentralized open ledger. I.e., transactions are verified and executed out on the blockchain, which is stored on multiple computers, instead of through a central authority (government or bank). Hence, transactions are recorded publicly to ensure counterparty trust and every user can track assets which are registered on the blockchain. Apart from that, it is supposed to be less costly than a central authority executing transactions since an intermediary is cut off.²To make this process more visible, consider a contract concluded through a purchase by credit card for instance. In this scenario, the credit card user buys a good from the seller but usually has to pay for an already included surcharge of the merchant, since he has to pay fees to the credit card company when accepting payment via credit card. Additional costs like this could be avoided when using the blockchain.³

To put it in different words, the blockchain is - amongst others - a technology to witness and carry out transactions instead of a notary or bank. Next to public blockchains, which anyone may read or edit, there are also private or permission-based blockchains where different users could have permission for different operations on the blockchain.⁴In these blockchain networks, a community of other users can control how the stored records of transactions (i.e., the blockchain) are amended or updated. As mentioned above, the blockchain is a decentralized network, hence not a single person controls the information. By replicating the ledger in multiple computers (nodes), the transaction history becomes publicly accessible and distributed. The Community can approve the transaction by checking the ledger (proof of work).⁵Proof of work is currently the most-used consensus mechanism and if the majority of the network votes to approve the transaction, it will be validated and a new block will be added to the chain.⁶This process requires solving a complex cryptographic puzzle which demands large amounts of CPU power because it is done by trial and error. Various specialized users (miners) compete in solving it since the Bitcoin blockchain for instance rewards the creation of a new block with Bitcoin. Since the newly created block contains a hash reference to the previous block, the structure is called blockchain.⁷By duplicating the blockchain on multiple computers and the concatenation of all following blocks, the system is furthermore protected against manipulation and unjustified amendments. Since each block contains a timestamp and a specific link to the previous block, unlawful amendments would lead to discrepancies within the blocks which are easily observable.

The modus operandi of the blockchain is especially ground-breaking with regards to payments. Traditionally, payment systems guaranteed security and proof of ownership by relying on central and trustworthy third parties like banks or credit card companies. Contrary to that, the blockchain solves this issue through a peer-to-peer solution. Hence, it is able to replace the third party because it contains the history of all previous transactions and users can easily track down which person owned something during a given time without uncertainty.⁸This system supposes that miners act out of self-interest and while doing so, fulfil a socially beneficial role by enabling the blockchain to carry out and approve transactions. In technical writing, some scholars divide the areas of application into three parts. First of all, the blockchain 1.0, which is heavily focused on the concept of cryptocurrency and the decentralization of money transfers mentioned above. This thesis is however mainly based on blockchain 2.0, representing the use of smart contracts. Lastly, blockchain 3.0 describes the connection between the blockchain and Internet of Things-technology and applications relating to health or governmental bodies (decentralization of the digital society).⁹The connection between smart contracts and the IoT will also be analysed later in this thesis.

2. Potential Threats

It should be mentioned that blockchain networks tended to shift from a dispersed community to a structure with fewer, but more dominating actors. This could be a distortion to the development of blockchain networks and smart contracts because the use of consensus mechanisms bears the risk that the further development will move towards the preferences of the majority and will be less focused on the interests of minorities. This development might actually be due to the rising popularity of the blockchain. When more transactions are carried out on the blockchain, the validation process becomes more difficult. As a result, fewer users with regular computers are able to mine blocks. Due to that dynamic, miners created so-calledmining pools, allowing them to combine their CPU and share the obtained rewards. One of the main concerns with this is, that a group with enough computational power might be able to block certain valid transactions. This problem is similar to the hypothesis of Mancur Olson, stating that subgroups (mining pools) could exploit larger groups (blockchain networks), resulting in a loss of social welfare. The potential risk of mining pools behaving in an opportunistic manner might create additional transaction costs.¹⁰

II. Smart Contracts

1. General Description

Originally, Satoshi Nakamoto’s vision of the blockchain only referred to monetary transfers, but transactions executed and recorded on the blockchain might contain any set of promises (e.g., a contract).¹¹These so-called smart contracts were first introduced by computer scientist and lawyer Nick Szabo and some call them the most disruptive application of the blockchain until now.¹²Hence, the expansion of smart contracts is developing fairly quick and during the last couple of years several projects, such as Ethereum or Mastercoin, have been founded to create programming languages which enable the formulation of increasingly sophisticated smart contracts.¹³Nevertheless, the name smart contract is somehow misleading since they are not actually smart in the sense of an artificial intelligence, but simply execute provisions which were agreed upon ex ante. Smart contracts are essentially computer programs that are executed by a network of nodes without the discretion of a trusted authority and represent the implementation of a contractual agreement. They are based on computational if-then logic, meaning that the set of rules of a smart contract automatically triggers predefined reactions corresponding to particular probabilities.¹⁴This empowers the parties to structure their relationship in a more efficient way without having to deal with the ambiguity of words.¹⁵Apart from that, they do not rely on external enforcers anymore which tend to be costly. In conclusion, smart contracts aim to merge the formation of the contract as well as its enforcement in one sole instrument.¹⁶

1.1 Illustration of the Code

To illustrate how a smart contract may in fact look like, imagine someone betting a certain amount of money on the fact, that a random number is smaller than 1000. In case the person is right, his stake triples but if not, the stake is lost. The code for this contract would look like this:

contract bet {

function() {

var stake = msg.value;

if (block.timestamp > 1000)

msg.sender.send(3 * stake);

else return;

Abbildung in dieser Leseprobe nicht enthalten

1.2 Visualization of a Smart Contract

Abbildung in dieser Leseprobe nicht enthalten

Figure 1: Visualization of a Smart Contract, Felix Schiller, May 2019

2. Platforms

In this section, this thesis shortly presents the most important platforms on which smart contracts can be implemented. There are also several other platforms offering similar services, but Bitcoin and Ethereum boast by far the highest market capitalization amongst them which is why this thesis describes especially those two.¹⁷To begin with, it will be analysed how these platforms might be perceived from an economic and legal perspective.

2.1 Economic and Legal Perspective of a Blockchain Platform

In an abstract manner, digital platforms are understood as IT systems which enable users and suppliers to conclude business transactions and to generate revenue. Apart from that, parties engaged in business activities on digital platforms, can also be advertisers or software developers. Most importantly, digital platforms are usually free to enter and publicly available. This leads to an innovation boost since entry barriers are lowered and basically everyone can participate in the development of different platforms, causing immense network effects. Referring back to the blockchain, it might be perceived as an enabling tool for the next generation of digital platforms. A blockchain platform deviates from the previous understanding of a digital platform insofar, as it is not run by a specific enterprise but by a multitude of parties, making it a decentralised platform.¹⁸In this context, the question arises which social and economic advantages the blockchain has compared to previous platforms such as databases as we know them so far. Generally, a blockchain platform can be compared to a distributed database. Like a distributed ledger, distributed databases are replicated across different nodes, leading to several advantages. These advantages include increased scalability and additionally better reliability and availability of the database. However, unlike it is the case with a traditional database, a copy of the data stored on the blockchain platform cannot be modified simultaneously, which leads to less conflicts than before. As already mentioned before, the data on the blockchain can be modified by transactions executed by a built-in virtual machine (e.g., via smart contracts).¹⁹

2.2 Bitcoin

First of all, there is the publicly well-known Bitcoin platform, whose primary goal is transferring currency. Yet, due to the openness of its blockchain, the platform also enables the implementation of smart contracts. However, its scripting language is rather limited and only a small part of its nodes is able or willing to process complex transactions.²⁰

2.3 Ethereum

2.3.1 General Information

Many blockchain experts share the perception that the Ethereum platform is better suited to process smart contracts than the Bitcoin platform. Essentially, the platform is also based on a consensus-algorithm but allows the use of the programming language Solidity, compiling into bytecode language. This enables users to write code for relatively complex smart contracts.²¹

2.3.2 The DAO Incident

When it comes to the downsides of Ethereum, it should be mentioned that a venture capital fund structured as a smart contract and based on the Ethereum platform was hacked in 2016 and $60 million worth of Ether were stolen.²²The fund was a so-called DAO (i.e., Decentralized Autonomous Organization) which is a crowd-sourced investment vehicle. In general, DAOs are open-source which means that they are transparent and, in theory – just like the blockchain - incorruptible. The aim of a DAO is to codify the rules of an enterprise or institution, by which it eradicates the requirement of supplementary documents and a governing management structure, creating an organisation with decentralized control.²³Referring back to the DAO incident in 2016, investors sent Ether to a mutual fund (i.e., ICO or Initial Coin Offering) and the code of the smart contract was supposed to allow the investors to vote on the fact in which way the collected cryptocurrency should be invested. At this point, it was the biggest crowdfunding in history. However, it should be noted that such tokens are not considered equity shares but more like contributions which give investors voting rights instead of ownership of the DAO.²⁴Apart from that, the smart contract provided an exit function to the investors, enabling them to withdraw their Ether. At first glance this made sense but due to a flaw in the code users were able to withdraw more crypto money than they originally invested. Since it is not possible to revert transactions on the Ethereum blockchain and off-chain remedies, such as litigation, were not possible due to the anonymous nature of the DAO, there were no possibilities to transfer the unlawfully withdrawn Ether back to the rightful owners.²⁵This led to a so-called hard-fork by the Ethereum founders in order to reverse the fraudulent transactions but therefore also denying the immutability of the smart contract. Since several crypto exchanges and miners did not agree with the Ethereum founders acting comparably to a centralized authority, they backed up the alternative currency Ethereum Classic (ETC). This currency utilizes the original blockchain and crypto investors holding Ether on it retained their rights. Later on, both Ether and ETC were able to position themselves in the market and ETC even became the third most traded cryptocurrency after Bitcoin and Ether at some point.²⁶The DAO incident was also one of the reasons why distributed ledger platforms started to oblige participants of the network to identify themselves in order to prevent similar fraudulent behaviour as in the DAO case, where no identification was necessary.²⁷

C. Potential Fields of Application

Due to their immunity to manipulation, the use of smart contracts is interesting in many scenarios. This applies especially to cases where simple money transfers are required to comply with certain stipulated rules. Generally, smart contracts can be used to execute two types of transactions. This includes the automatic execution of a contractual payment obligation if the counterparty fulfils its condition or an automatically executed penalty if the contracting party does not perform its contractual obligation.²⁸Currently, the evolution of the blockchain is still in an early stage and it will need time until code is able to set out more subjective legal criteria. This may include the determination of whether a party did satisfy a commercially reasonable efforts standard or if an indemnification clause should be triggered for instance. However, one of the requirements for this development is, that more assets will be tokenized in order to become transferable through the blockchain.²⁹The examples below are comparably ordinary in general and thus they seem to be suitable to be carried out in the form of a smart contract. Whether this turns out to be true will be evaluated in the subsequent section.

I. Financial Services

Generally, financial institutions work with standardized terms and variables which are measurable. Hence, they usually do not need to rely on human judgement and terms that are ambiguous. Since smart contracts are designed to be rigid and do not leave room for interpretation, this is a tremendous fit. Until now, smart contracts have been particularly used to execute financial products such as derivatives, futures and options. In addition, they could be used for trade clearing by managing the approval workflow between the parties and also transfer funds once the trade settlement amounts are calculated. With regard to crowdfunding, smart contracts could track the number of funds submitted by investors, and as soon as the amount reaches the funding goal of the campaign, the money will be transferred to the initiator of the campaign. In case the goal will not be reached within the deadline, the funds will be automatically returned to the investors.³⁰One of the reasons why the blockchain might be capable to disrupt the financial services sector, is the fact that financial activities mostly used to rely on convoluted accounting systems and an enormous bureaucratic apparatus verifying and transferring an extensive amount of data on a daily basis. Until now for example, swap trades are rather complicated and contain a master agreement, a certain schedule and other optional documents which still have to be executed manually and sometimes even require a signature provided via fax or scan to lead to a transaction confirmation.³¹A decentralized ledger as the blockchain is considered as a true game-changer in this context since it is able to process financial transactions like this as well, but it imposes fewer costs on the parties and is unequivocally faster than the traditional system. Especially international transactions tend to be costly and inefficient, hence blockchain technology would certainly be serviceable in this sector.³²

II. Insurance

Another interesting field of application is the insurance sector. Smart contracts could transfer the insurance payment to the insured party as soon as the pay-out is determined based on the type of claim and the underlying policy which is encoded in the smart contract.³³For instance, this is discussed with regard to travel insurances, where a smart contract based on the Ethereum blockchain could automatically reimburse travellers if their train or flight has been delayed. Thus, the adoption of blockchain technology and smart contracts would increase the speed of processing claims and reduce costs due to the manual processing of claims. Nonetheless, it should be mentioned that smart contracts in their current state can only be used for a limited amount of insurance policies. In fact, a greater proportion of claims still have to be assessed by a human expert before being resolved. Regarding the structure of the applied blockchain, it would make sense to combine a private with a public blockchain. In this case, the private blockchain could record policies and claims, whereas the latter could execute the refund by transferring crypto money to the wallet of the claimant. Another option might be to solely exploit a public blockchain. This would increase the trust level of customers since the insurance process would be entirely decentralized, but would also impose higher transaction costs on the parties due to dependence on external miners.³⁴

III. Healthcare

Lastly, smart contracts might also enable the access to the health record of a person as soon as the multi-signature approval is established between the respective providers and patients. In this respect, medical researchers could gain access to the health data of users by paying a small amount of money to the patients for their participation and disclosure of their data. In the era of smartwatches and IoT-devices that track the health of a person, health insurances could also reward customers by paying them a certain reward when they hit a specific goal like walking 200km in a month.³⁵However, it is questionable whether the storage and potential distribution of such sensible information is compatible with current regulation on data and privacy protection.

IV. Conclusion

In conclusion, the adoption of the blockchain technology is not limited to a single business sector or scenario. Moreover, it may benefit different kinds of contracts or businesses, but it remains questionable whether the enthusiasm of many blockchain pioneers and experts is justified considering the complexity inherent to some of the envisioned areas of adoption.

D. Compatibility of Smart Contracts with Complex Contracts

In this section, it will be evaluated if and eventually how smart contracts can be used for convoluted transactions that are more sophisticated than just simple payment transactions through the blockchain. In general, all contracts are based on conditional statements which can be translated to computer code.³⁶Nevertheless, there is legitimate doubt whether this technology truly makes legal and economic sense for complex contracts. First of all, this thesis briefly discusses how a smart contract can be legally qualified. Afterwards, it will analyze language barriers, behavioural challenges especially with regard to incomplete contracting and difficult set-ups concerning the contracting parties. Furthermore, I try to assess and develop solutions for the application of smart contracts in a more difficult environment of contracting.

I. Definition of a Complex Contract

As I did mention in the introduction and the first section of this thesis, several blockchain pioneers continue to promote the blockchain technology and smart contracts as a disruptive mechanism for contracting and the way business is done in the future. However, the question arises whether this does only apply to the way simple financial transactions are carried out or also to complex contracting. Hence, I am trying to set out a distinction for this matter in the subsequent section. First of all, a definition of a sophisticated, complex contract is necessary here. It seems as multiple blockchain enthusiasts tend to classify contractual agreements merely as financial transactions. With regard to such agreements, it undoubtedly makes sense to optimize them by using the blockchain technology. This is because ambiguity and contractual flexibility do not play a large role with less complicated transactions. Since simple financial transactions usually do not lead to severe legal disputes, judicial remedies are probably not required as well. In addition, enforcement costs are reduced due to the self-executing nature of smart contracts. Thus, I share the opinion of the experts heralding smart contracts until this point. Anyhow, it should not be forgotten that contracting is about much more than just reducing transaction costs and automatize as much as possible.³⁷Complex contracts consist out of several obligations but especially in those set-ups, things often do not end up according to the original plan.³⁸An efficient contract however should anticipate these issues and try to resolve them ex ante. To explain the issues of a complex transaction, it might also make sense to elaborate on Shavell’s definition of a production contract (or a contract governing the performance of a service). In his view, several uncertainties come along with these types of contracts. This includes the fact that the buyer might not be sure about the value of the purchased good and that the seller might be uncertain about the production costs at the time they are concluding the initial contract.³⁹If such a production contract would be governed by a smart contract, these variables could probably not be amended anymore after the contract is concluded. This would make contracting possibly less attractive for parties not willing to fix every term of the agreement ex ante. In addition, there are several other traditional contracting conducts, potentially making the contracting process too complex with regard to the nature of a smart contracts. This will be analysed subsequently in this thesis.

II. Legal Perception of Smart Contracts

Various scholars are debating whether a smart contract can be legally classified as a traditional contract and there is still no universal legal definition of a smart contract. Traditionally, a contract is recognized as a commitment to the terms the parties agreed upon and also as a physical document which contains these written rights and obligations itself.⁴⁰In contrast, a smart contract is partially perceived as a self-help mechanism because its self-executive nature leads to the expendability of courts, thus some consider it not a contract itself. However, it appears that smart contracts might be more than just an instrument for self-help since they are not only a mean to execute the contract, but also contain the terms of the contract itself or might serve as an escrow account.⁴¹Furthermore, it certainly depends on national law, in which way a smart contract might be legally classified in the end. Under German law, some experts consider the code of a smart contract not as a written contract document that is agreed upon by the parties and by which they shall develop a legal relationship. Instead, it is simply supposed to execute what is set out by a regular contract. However, this execution may coincide with the conclusion of a legal obligation. It is questionable, whether the contracting parties can use the program code to formulate legal relationships (comparable to a written contract document) and to express the content of their declarations of intent. Nevertheless, it is assumed that this might be at least the case for smart contracts which are set out on an individual basis since they shift legal rights and responsibilities similarly to an agreement between two parties who physically exchange commodities for payment over a counter.⁴²Although considering the areas of application of smart contracts, it can be detected quickly, that individually negotiated smart contracts are presumably an exception. Usually, smart contracts are pre-formulated for multiple legal relations and hence, the restrictions of the so-calledAGBlaw (i.e., law on terms and conditions) are most likely to apply in Germany.⁴³

II. Fees and Capacities of the Blockchain

Another difficulty that is mentioned regularly in blockchain literature, when it comes to the discussion whether smart contracts can be used for complex transactions, is the fact that they may impose extensive fees on the contracting parties. In order to execute a smart contract on the blockchain, a transaction fee must be paid which is known as “gas” on the Ethereum blockchain for example. However, if the transaction steps of the smart contract are relatively intricate, a higher fee has to be paid to execute the contract.⁴⁴Contrary to that, fees on the Bitcoin platform are paid on a voluntarily basis, but this might cause tragedy of the commons problems. Currently for instance, too many transactions are taking place on the Bitcoin blockchain, which is why it takes relatively long to execute them. Unequivocally, transaction fees in general are justified since every transaction on the blockchain imposes costs on the network due to the requirement of download and verification. In this context, they serve as a regulatory instrument to prevent over-use or even abuse.⁴⁵Nonetheless, there are concerns whether public blockchains can be used for daily contracting since the process might become too slow if more and more people will switch from traditional to smart contracts. That being said, it should be mentioned that there are projects under development which might be able to solve the scalability issues of the blockchain, such as the Lightning Network for the Bitcoin platform and the Raiden Network for Ethereum. The main idea behind these projects is to combine online and offline transactions to reduce mining costs and to fasten the execution process.⁴⁶

III. The Language of the Code

As already mentioned above, smart contracts are not written as natural text but computer code. Hence, they are understood by the computers running them, but not by ordinary people or even the majority of lawyers. Especially with regard to complex contracting, this may create difficulties which have to be solved in order to establish smart contracts as an equivalent alternative to traditional contracts.

1. Semantic challenges

Usually, the parties of an agreement are non-programmers, unlikely to fully understand the context and consequences of the coded contract and far from being able to create one by themselves. Even though a user-friendly interface may cover up the technicalities, the inherent language barriers cannot be fully avoided.⁴⁷This may not be a problem for simple agreements without any severe consequences, but for complex and important transactions it most likely is. As a result, potential users of smart contracts are forced to seek advice from experts which are able to code on the one hand, but also can understand the legal side of the contract on the other, which is rather costly. In addition, they indeed get rid of a central intermediary but still need to rely on programmers or other service providers helping them to implement and understand the code. From an economic point of view, transaction costs are simply shifted from the enforcement and monitoring spectrum, as it is the case with traditional contracts, to costs resulting from the consultation of third-party experts like programmers or specialized lawyers. In addition, there are concerns about whether courts are able to review and interpret the code because of its technicality and lack of ambiguity. Besides, computers are currently not capable of understanding complex contracts which are expressed in natural language. Although these semantic obstacles might be just temporary, they lead to several concerns regarding the practicality of the combination of law and code at present. In conclusion, within the context of its understandability, smart contracts are currently not an efficient solution for agreements requiring complex negotiations. Especially in the case of a complex contract being in the centre of an eventual lawsuit, judges probably will not understand it adequately. Nevertheless, it does make sense to use smart contracts for standardized agreements like derivatives contracts because their scalable application outweighs the costs of consulting expensive programmers and advisors.⁴⁸

2. Ricardian contracts

While considering solutions for the language barriers of smart contracts, so-called Ricardian contracts should be discussed at this point. Just as smart contracts, they were invented during the late 1990s and were intended to be comprehensible for software but also regular users and the legal system. This is possible because - unlike a smart contract - the contract itself is formulated in natural language but contains machine-readable tags and can be signed digitally. The agreement is encrypted through the use of hashes, enabling the creation of a message digest which serves as an instrument of identification and evidence of intent. The message digest is used in the first transaction and subsequently in every transaction during the execution of the agreement. While smart contracts are mainly focused on the execution itself, Ricardian contracts try to be as much comprehensible for non-programmers as possible.⁴⁹

3. Smart Contract Templates and Domain Specific Languages

Another approach to make smart contracts more comprehensible could be the implementation of standard templates. These templates are supposed to establish a legal framework especially with regard to financial transactions. Furthermore, the intention is to develop and use a domain specific language which enables the execution of various functions such as the potential execution of the agreement on several platforms. However, the concept of domain-specific language is not entirely new. Domain-specific languages are designed for a specific application or domain and have only a small variety of features which are perfectly suited for their intended use. Because of this, they seem to be a good fit to code smart contracts. An example of a domain specific language is the previously mentioned Solidity, created for the Ethereum platform. To make smart contracts more comprehensible, domain-specific languages could also be used to create a modelling platform for smart contracts. The graphical interface of such a platform is supposed to allow non-programmers to draft smart contracts. If a certain test run of the contract turns out to be successful, the contract can be deployed to the relevant blockchain network.⁵⁰

4. Declarative Smart Contracts

Another promising solution for the semantic problems that come along with writing a contract as code, might be the application of declarative smart contracts. Until now, smart contracts are mostly formulated in an imperative language, meaning that they state the computational operations to be carried out in a procedural language and hence are hardly comprehensible for the average contracting party. In addition, the order of the instructions given in the code has an effect on the result of the smart contract. This leads to the fact that the process of drafting such a contract is prone to errors. In contrast, declarative smart contracts are closer to natural language. The general idea behind declarative language is that the coder is not forced to set out in detail how the contract has to be carried out. Instead, the creator simply needs to state what the contract should do in the end. Such a contract can be written in logic-based language representing the contractual terms. Because of this logic-based approach, the given statements generate further statements, which in the end lead to the desired performance of the contract. To put it in other words, a declarative smart contract is bound to set out the semantics to be employed, while an inference system is able to reason up the contract and execute it. Nevertheless, the costs of an imperative contract are lower because they require less computational complexity than a declarative smart contract. This is presumably one of the major reasons why smart contract programmers have been mainly using imperative languages thus far. In addition, it seems problematic to install off-chain inferential mechanisms on-chain. However, this might be possible if a so-called oracle is available, which will be explained and analysed in a subsequent section of this thesis. Notwithstanding, the declarative smart contract could be executed fully off-chain, while its content and inferential conclusions could be recorded and stored on the blockchain. Afterwards, the imperative code derived from these conclusions could execute specific transactions on-chain.⁵¹However, digging deeper into these sophisticated technicalities goes beyond the scope of what this thesis tries to elaborate on.

5. Legal Enforceability of Programming Language

Lastly, questions also arise regarding the legal enforceability of a contract which is not written in natural language. Many jurisdictions such as in the US or France require a contract to be certain and somewhat “clear”. As already pointed out, non-programmers are not able to understand computer code thoroughly. Thus, judges for instance could have problems interpreting the code and hence may not regard it as certain. What is more, even if a natural language version of the smart contract would exist, the so-called parol evidence-rule could be problematic in the US. This rule sets out that potential evidence outside of the main contract (i.e., the natural language version) should not be considered within the interpretation of the contract. This may not apply, if the terms of the smart contract are in fact fully incomprehensible and only the natural language version can be interpreted. Nevertheless, in either scenario it seems likely that the interpretation process is slowed down since there are two versions of the contract to consider.⁵²It also might be possible that terms of the contract written in natural language might not have an equivalent in the code of the smart contract or the other way around. E.g., this could apply to a clause in the traditional version governing reasonable care since the computer code cannot properly deal with ambiguous terms like this. Unfortunately, there is no useful case law for smart contracts so far and hence an adequate fact-based conclusion on this issue cannot be drawn here.⁵³Nevertheless, scholars state that the enforcement costs will likely depend upon the fact, to which extent human discretion might influence the transaction.⁵⁴

IV. Incomplete Contracting and Behavioural Challenges

As mentioned before, the goal of contracting, especially with regard to more sophisticated agreements, is not only to reduce transaction costs. Sometimes it also serves broader social purposes (i.e., setting expectations for future behaviour or facilitating stable but flexible long-term relationships), and there are severe doubts that smart contracts are capable of fulfilling them. In addition, traditional contracts are used to rely on contracting standards such as the duty of good faith, which might not be properly implementable into code.⁵⁵

1. Inclusion of Non-enforceable Terms in Contracts

With regard to traditional contractual behaviour, it could be observed that parties tend to include terms in their contracts which might not be enforceable in front of a court (e.g., terms in a rental agreement which would lead to unlawful contractual disadvantages of the tenant). At first sight, it seems irrational to include them and therefore irrelevant whether smart contracts may be able or not to comprise such non-enforceable terms. However, the counterparty might interpret such terms as lawful and thus stick to them anyway. Due to this potential imperfect knowledge, such described contractual behaviour can deter conduct beyond what is actually intended by the law. Therefore, for one of the parties it might pay off to include such terms in the contract. Despite that, it should be mentioned that this conduct can in some cases even serve both sides of the agreement. As an example, consider a situation where both parties agree to implement a clause which is presumably non-enforceable but which might still affect their future behaviour, potentially making them stick to the terms agreed-upon (e.g., “infidelity clauses” in prenuptial agreements).⁵⁶

2. Inclusion of Vague Terms in Contracts

Comparable to the behavioural phenomenon above, parties also tend to include terms in - especially long-term – contracts which are fairly vague and leave certain issues intentionally unanswered. This practice exists mainly to lower negotiation costs, but also because parties might want to leave some space for interpretation, making the contract more flexible in the future due to its vagueness. In addition, they also rely on their business relationship outside of the scope of the written agreement.⁵⁷

3. Intentional Non-enforcement

Lastly, in some cases parties might have a legitimate claim against their counterparty but still refrain from enforcing it. This conduct can be due to their intention to remain on an amicable basis with their business partner and thus use the law as an instrument to manage their interpersonal relationship. It is also based on the fact that litigation tends to be costly as well as time consuming. By using the judicial system simply as a backdrop or leverage, the chances of reaching a mutually beneficial settlement outside of the court system might be increased (e.g., “bargaining in the shadow of the law”).⁵⁸Contrary to the outcome of a judicial trial which is typically rather one-sided, this conduct of non-enforcement provides for a solution which might be more cooperative and balanced in the end. Naturally, this practice more likely leads to a successful outcome, if the agreement was concluded in a network with strong social norms and effective signalling mechanisms (e.g. online reviews or, more traditionally, gossip), which serve as an instrument to force the parties to remain on an amicable basis.⁵⁹A self-executing smart contracts however does currently not allow this practice which may lead to undesired socioeconomic results.

4. Efficient Breach

Another concern regarding the compatibility of smart contracts with traditional contract theory is the existence of efficient breach. In the case of efficient breach, non-performance of the contractual obligation is economically more efficient.⁶⁰This may be due to the fact that circumstances might change after specifying the contract, leading to potentially unwanted economic or legal effects (e.g., higher production costs than expected, making it unattractive for the seller to perform his contractual commitment). A smart contract however is fixed, hence it may lead to suboptimal outcomes in case performance is not desired by either one or both of the parties. Numerous law and economic scholars share the opinion that contract law is geared towards enabling such non-performance since they assume that the legal default rules for contractual remedial measures are part of the parties’ contract negotiation.⁶¹To enable efficient breach in smart contracts, the parties could implement terms of force majeure. However, as already mentioned in this thesis, such imprecise clauses do not work well together with the definite nature of code. In addition, contrary to traditional contracts, smart contracts cannot be modified easily. Although a smart contract could incorporate an option to modify the terms before its execution, this would have to be determined explicitly in the contract ex ante and would further increase the complexity of the contracting process.⁶²

5. Reduced Legal Uncertainty

One of the drawbacks of incomplete contracting is the inescapable legal uncertainty. As Posner mentioned in his paper about the law and economics of contract interpretation, uncertainty due to interpretive problems of courts may arise even in a setting of perfect foresight. Since a traditional contract does not provide a solution for every potential outcome of an agreement because parties may rationally decide to economize negotiation costs, both of them form an expectation about what a court might eventually decide. This is particularly the case when the parties assume that there is just a minor probability that the relevant event will materialize.⁶³As a result, their expectation will then be accounted in the price of a service or good upon which the parties agreed (e.g., costs of insurance regarding the risk from court interpretation for a party which is risk averse). Hence, if both parties are risk averse, they might not be able to negotiate a mutually beneficial price. In addition, the parties might decide not to contract in the end if their expectations of a potential judicial decision diverge excessively. This is the case when the expected costs by the seller of supplying the good are critically higher than the expected value for the buyer (i.e., the anticipated probability of getting the good times its value). In this regard, a smart contract could be able to minimize the risk of uncertain judicial decisions since a computer interprets the coded contract like a strictly textualist judge would. Therefore, the outcome of the agreement would be more foreseeable as it is the case with the uncertain interpretive methodology of a court.⁶⁴As it is the case with reduced counterparty risk, reduced interpretive risk may lead to lower prices since the risk premium of a risk averse party of the contract is lower. The higher predictability of smart contracts could induce parties to trade a good or service which would not be traded if a traditional contract is used and thus resources might be allocated more efficiently. In this regard, smart contracts could reduce social costs.

6. Modification of the Contract and the Hold-up Problem

Similar as with efficient breach, a change in the environment of the contract after its conclusion may cause the desire of the parties to change specific terms. As mentioned above, it might be advantageous for the seller to breach the contract if the costs of his performance have risen. In such cases however, the buyer could still offer to shift some of his contractual surplus to the seller to keep the contract alive by modifying it (i.e., Coasian renegotiation).⁶⁵This is generally desirable, but such ex post modifications may also be the result of the hold-up problem which is a specific type of transaction costs. Imagine the buyer made a relationship-specific investment which has a higher value to the contract parties compared to a random third party. To benefit from this situation of the buyer, the seller might breach the contract if the buyer does not agree to a higher purchase price. Since the buyer already invested in the outcome of the contract and cannot fully recoup this investment, financially it might make more sense for him to pay the increased price.⁶⁶This so-called hold-up problem has prevented parties in the past to invest in agreements even though relationship-specific investment could be economically efficient. When it comes to enforcing such modifications, it has been challenging for courts to distinguish between the two cases described above where only the first scenario is desirable.⁶⁷Scholars claim that parties do not sue in these hold-up scenarios when legal remedies are inadequate. This is mostly due to high litigation costs and the uncertainty of the judicial outcome but also because the law does not provide efficient remedies for specific types of hold-up. In the past, contracts theorists developed mechanisms to solve this problem, but they require a commitment of the parties which is hard to achieve with the present contracting techniques (e.g., no renegotiation clause or renegotiation-design mechanisms to maintain some flexibility but they require ex ante information by the parties which is usually not given). In addition, information asymmetry has been an issue which prevented overcoming the hold-up problem. Smart contracts however could enable such commitments by implementing a penalty clause which is triggered when there is a deviation from the original contract or mechanism. Such a penalty will then be paid to third parties on the blockchain without any interference by courts or the parties.⁶⁸Accordingly, the parties cannot renegotiate as long as they are not able to make to make side payments that are verifiable by the code which makes them stick to the initial agreement. The question whether this mechanism indeed works in practice, still needs to be answered but nevertheless it could be an improvement to current contract law.

7. Conclusion

Due to their definite and self-executing nature, it seems most likely that smart contracts do not harmonize with these kinds of contracting practices and cannot take the social complexities of contracting entirely into account.⁶⁹Without the possibility of being used as a social tool to manage long-term relationships, smart contracts appear to be less attractive for complex contractual agreements. Obviously, they may create efficiencies but as the issues set out above underline, it seems as the benefits of smart contracts vanish, considering the existing social customs when contracting in real life. This applies particularly with regard to personal services, where the parties trust each other and desire to engage in a continued relationship beyond the initial contract. In these cases, the self-enforcing nature of smart contracts is arguably an inferior option compared to the self-executing social mechanisms such as reputation and relational stability.⁷⁰However, it should be mentioned that the proclaimed vagueness of incomplete contracting as it is known until now, may also lead to contractual opportunism which is not necessarily desirable. This opportunism can be solved by smart contracts due to their transparency and self-executive nature. In conclusion, both traditional and smart contracts have advantages and disadvantages compared to each other. This being said, it appears - as of today - that smart contracts are not a serious threat to traditional contracts because firms and individuals still mainly rely on incomplete contracting since they generally impose lower transaction costs, whereas the blockchain technology solely works with complete contracts. However, from a behavioural economics point of view, smart contracts could eventually lead to more efficient transactions since concerns due to adverse selection (ex ante) and moral hazard (ex post to the transaction) are reduced (i.e., because of the self-executive nature, costly signalling can be avoided).⁷¹

V. Privacy and Information Disclosure

1. General concern

Another issue worth considering when contracting through the blockchain technology is the potential loss of business privacy due to the evident disclosure of information. On the Bitcoin blockchain for example, all transaction details except the identities of the transacting parties are publicly available information. Nonetheless, for some contracts (e.g., insurance agreements) even the disclosure of identity seems necessary. Especially financial institutions tend to be careful however when it comes to publicly disclosing transactional details. Nonetheless, there are several solutions for this problem, aiming to hide sensitive information by improving the encryption of the transaction. For instance, if the verification of the transaction would require a smaller number of nodes approving it, privacy might be increased but this would also negatively affect the trustworthiness of the verification process.⁷²Alternatively, a certified intermediary could verify the identity of users when they use a specific service for the first time and subsequently link it to their address. Thereafter, further identification will no longer be necessary when these users sign a smart contract and they only need to disclose their credentials. However, the credentials could be stolen or lost and because the blockchain does not work with intermediaries, these credentials cannot be reset. This could be solved by working with external service providers, which can store the credentials and return them to the user if they are stolen or lost. This being said, the adoption of such services bears drawbacks as well since they gain access to sensitive information of the users.⁷³

2. Theoretical Examples of Information Disclosure Challenges

Imagine an exporter selling rare meat to a restaurant chain, where the parties rely on a logistic service company and ports for shipping, as well as on intermediaries financing the transaction. By making use of an IoT-sensor, the parties could monitor the temperature, location and final delivery of the goods. The remaining participants on the blockchain may receive this kind of information as well in order to monitor the process. During this process, the final information is not yet implemented on the blockchain. Hence, a decentralized consensus has to be created, determining whether the goods were delivered successfully and eventually enabling the recording of the input on the blockchain. To verify the delivery of the transaction, seller and buyer, logistic providers and record keepers such as other customers or other sellers have to be contacted, submitting their reports to the blockchain. This is due to the fact that if the required information would only have to be verified by the buyer and seller, the use of the blockchain technology would not add any value in this scenario. Subsequently, the blockchain protocol creates a decentralized consensus, thus adding a new block to the blockchain, which has to pass consistency checks regarding the history of the existing blockchain. Generally, it can be stated, that the chance of a high-quality consensus is greater if the verifying parties have as much information as possible. Hence, there exists a trade-off between the scope of disclosed information and the quality of verifying consensus mechanism.⁷⁴

3. Conclusion

Simple transactions like general payments on the Bitcoin blockchain do not require detailed information since the only concern with this is basically the risk of double-payment. When it comes to complex transactions, this does not apply anymore and off-chain information such as shipment status need to be disclosed to the record keepers to generate a verifying consensus. However, record keepers might be incentivised to report falsely since they could be part of the transaction themselves as mentioned above. This might be the case if the costs of misreporting are lower than the utility the record keeper gets in case of a wrong consensus.⁷⁵

VI. Contracts Relying on External Services

Another challenge that comes along with the rising complexity of smart contracts is the requirement of information or data outside of the scope of the blockchain and its implementation into smart contracts. Originally, the blockchain was not designed to input data from the world outside of it. However, to determine whether a payment condition was fulfilled, some smart contracts need to rely on communication with the physical world. This applies to so-called off-chain events, whereas on-chain events occur in the blockchain and hence are visible to it.⁷⁶

1. Oracles

To overcome the problem described above, so-called oracles might be helpful. These are interfaces, which provide data from an external source to the smart contract.⁷⁷Generally, smart contracts can be unlocked and exercised by inserting private keys which only can relate to on-chain events. Therefore, an oracle, which should be a trusted third-party, has to be involved if the smart contracts requires off-chain data. Oracles can digitally sign the unlocking script to enable the transaction after verifying whether the relevant off-chain event occurred or not if a smart contract conditions the pay-out on its occurrence. However, it might be problematic that the oracle cannot evaluate the contractual performance properly since it only does or does not sign the unlocking script.⁷⁸

1.1 Oracles and Public Information

Software oracles are able to obtain the relevant information from commercial providers like Bloomberg or other blockchains if the event is based on public information like stock prices or the weather report.⁷⁹However, in some cases this does not guarantee that the event did in fact occur. Therefore, smart contract users may need to not only agree on an oracle but also on a trustworthy provider of information. In addition, since both the oracle and the source of information might be false or could be manipulated in the end, it is advisable to rely on a multitude of oracles and therefore different sources. Take for example an agricultural insurance policy that will pay out to the customer based on the received rainfall. Likely, the oracles will have to retrieve the required weather information from an external source. In this case, the information obtained by the different oracles may deviate from each other. Therefore, the parties should also ex-ante agree upon a percentual threshold of oracles, confirming the triggering event in order to sign the unlocking script and as a result, carry out the transaction.⁸⁰Apart from that, some firms have developed oracles that are able to certify the authenticity of the extracted data for a small fee.⁸¹

1.2 Oracles and Non-public Information

For such events described in the paragraph above, the required information is rather easily obtainable since it is public. It becomes more complex however if the respective event or information is not publicly tracked. This does apply to the arrival of a delivery package for instance, which may be supposed to trigger the payment by the customer via the smart contract. In these cases, the parties might need to establish sources of information by themselves (i.e., hardware oracles).⁸²Referring back to the delivery of a package, QR-codes or sensors could be attached to the box in order to establish a trustworthy tracking system. As soon as the box arrives at the customer, the oracles will be informed and initiate the pay-out process. Currently, this seems rather complicated and inefficient but with the rise of the Internet-of-Things technology (i.e., the extension of Internet connectivity into physical devices and regular objects providing real-time information about their state), it will become less costly.⁸³Moreover, IoT-sensors might be able to track the location of a package, but it seems questionable whether they could also measure the quality of a shipped product. If so, this would prevent the seller from shipping a low-quality product and actually solve a remaining issue when it comes to transactions where buyer and seller are not familiar and hence do not fully trust each other. Currently however, supply chains are not this sophisticated and therefore sensors cannot be used as the sole verification mechanism of a transaction.⁸⁴

Another issue with this technology is however, that one side of the contract may be digitalized but the other side still relies on some sort of alterable, physical conduct. Imagine for example a smart contract where a service provider is required to mow a lawn. As soon as an IoT-sensor verifies the required average length of the grass, the respective fee is transferred to the wallet of the gardener. Obviously, the latter part of the contract is executed automatically, but this does not apply to the alterable human behaviour of mowing the lawn. Hence, it might be possible that a court could still excuse the performance and thus the contract will not necessarily be executed.⁸⁵

1.3 The Machine Learning Solution

Another way of verifying a complex transaction might be the application of machine learning. Such algorithms are based on artificial intelligence and could take the data of IoT-sensors as mentioned above into account in order to predict the probability that the required performance of the respective party was carried out. In case the calculated probability is sufficiently high, the payment would be executed. In some way, the machine learning algorithm would replace a court, which has to consider whether the transaction took place adequately. This may reduce costs of potential disputes, but there are legitimate doubts whether contracting parties would trust a verification mechanism only based on machine learning.⁸⁶In addition, despite the innovative progress regarding machine learning solutions, computers are currently not able to deal with the degree of contextual, domain-specific, sophisticated complexity of contractual ambiguity. Furthermore, relevant smart contract platforms at the moment lack way behind advanced AI systems such as Watson which was developed by IBM.⁸⁷

1.4 Conclusion

Although oracles enable the implementation of off-chain data, they deviate from the basic incorruptible nature of the blockchain as Satoshi Nakamoto had it in mind. The contracting parties should also take into consideration that the oracle might stop to provide data at some point since it could be prone to technical issues or simply close down its service. This is why it is advisable to agree with the oracle on a minimum period of service.⁸⁸Apart from that, the implementation of Internet-of-Things devices will likely become important in the future, but as of today, the technology is still insecure and not necessarily scalable due to its lack of standardization.⁸⁹

VII. Termination of a Smart Contract

On a final note, another issue with smart contracts is their termination. In general, they are supposed to always execute themselves but especially with convoluted agreements, circumstances might change and thus the intent of one of the parties to finalize the transaction could shift as well. Smart contracts are indeed blocked if the respective wallet of the purchasing party is empty and no money can be transferred to the other party. Nonetheless, it appears that there is no termination mechanism if a party is unsatisfied with the content of the contract, supposes that the contract is invalid or believes that the counterparty has failed to comply with their obligation. As a result, the programmer of the contract would have to consider every state reached by the parties as long as the contract is in force. Apart from that, errors in the code could lead to unsatisfying outcomes which cannot be revised. It is doubtful whether programmers are able to work this flawlessly and anticipate every potential problem in the future. This being the case, there are certain interfaces which could prevent the undesirable execution of a smart contract by basically destroying it. Another solution for this issue might be to run the smart contract fully off-chain.⁹⁰The useful advantages of the blockchain technology would however logically disappear in such a scenario.

E. Conclusion

Considering all the facts described and explained above, it can be stated that there are several issues when it comes to the legal and technical boundaries of smart contracts and it is still questionable whether the technology is indeed designed to process complex transactions. Although the Bitcoin blockchain guarantees immutability and security, it currently does not possess the abilities to carry out complex transactions. Nevertheless, there are ways to carry out more sophisticated transactions, but these platforms or means are likely to be less immutable and decentralized. It can be stated that smart contracts still involve traditional inefficiencies and insecurities when it comes to contracting on a complex level, but will almost definitely change the way simple transactions are processed. It also has to be taken into account, that not every contractual obligation can be expressed in computer code. This limits the range of transactions which may potentially be carried out through smart contracts. In detail, this especially applies to contracts based on legal concepts such as “reasonableness” or “best efforts”, which require a more thorough and - as of now - human evaluation of the performance. This problem could be eventually solved by artificial intelligence implemented in the smart contract, but this will require further innovation and time.⁹¹On the one hand, the abstinence of human judgement lowers transaction costs and leads to less ambiguity, but on the other hand this may cause situations where the parties are not capable anymore of choosing whether and how they want to exercise their contractual rights.⁹²In addition, transactions costs are in fact only lower if the code does not include any flaws and exactly represents the intent of the parties which may not always be the case. In conclusion, smart contracts are likely to reduce the risk of non-compliance and human errors but still bear the risk of error in the code and may involve discrepancies between the original intention of the parties and their actual implementation in the contract, whose outcome is not easily reversible. The risk of error in the code rises particularly if the transaction is complex. Taking all the facts mentioned in this thesis into account, it can be stated that smart contracts do not lead to an immediate improvement of complex contracting in general. Nevertheless, jurisdictions establishing the most blockchain-friendly regulations are likely to have competitive advantages in the future when it comes to attracting new disruptive business models and enterprises which are willing to utilize them in a legal way. The further development of smart contracts does not only face technical challenges (e.g., standardization, scalability, security, 51-percent-attack, resource consumption, enormous amounts of data), but also depends on the solution of the lingering legal questions. Complex issues such as regulation, data and consumer protection law, liability and issues of law enforcement are only to be mentioned as examples.⁹³

References

Arruñada, Benito(2018), Blockchain's Struggle to Deliver Impersonal Exchange, Minnesota Journal of Law, Science & Technology.

Bartoletti, Livio; Pompianu, Massimo(2017), An empirical analysis of smart contracts: platforms, applications, and design patterns, Lecture Notes in Computer Science.

Bashir, Imran(2017), Mastering Blockchain, Packt Publishing Ltd.

Buterin, Vitalik(2013), A next-generation smart contract and decentralized application platform, White Paper.

Cieplak, Jenny; Leefatt, Simon(2017) Smart Contracts: A Smart Way to Automate Performance, Georgetown Law Technology Review 417.

Cong, Lin William; He,Zhiguo (2018), Blockchain Disruption and Smart Contracts, NBER Working Paper No. 24399.

Cooter, Robert; Marks, Stephen; Mnookin, Robert(1982), Bargaining in the Shadow of the Law: A Testable Model of Strategic Behavior, Journal of Legal Studies 11(2): 225-251.

Cuccuru,Pierluigi(2017), Beyond bitcoin: an early overview on smart contracts, International Journal of Law and Information Technology, 25(3), 179–195.

Davidson, Sinclair; de Filippi, Primavera; Potts, Jason(2016), Economics of Blockchain, Public Choice Conference, May 2016, Fort Lauderdale, United States.

Gans, Joshua(2019), The Fine Print in Smart Contracts, NBER Working Paper No. 25443.

Gatteschi, Valentina; Lamberti, Fabrizio; Demartini, Claudio; Pranteda, Chiara; Santamaría, Víctor(2018), Blockchain and Smart Contracts for Insurance: Is the Technology Mature Enough?,Future Internet10, no. 2: 20.

Giancaspro, Mark(2017), Is a ‘smart contract’ really a smart idea? Insights from a legal perspective, Computer law & security review 33 (2017) 825–835.

Governatori, Guido; Idelberger, Florian, Milosevic, Zoran et al.(2018), On legal contracts, imperative and declarative smart contracts, and blockchain systems, Artif. Intell. Law (2018) 26: 377.

Hart, Oliver(2017), Incomplete Contracts and Control, American Economic Review 2017, 107(7): 1731–1752.

Holden, Richard T.; Malani, Anup(2019), Can Blockchain Solve the Hold-up Problem in Contracts?, NBER Working Paper No. 25833.

Karamitsos, Ioannis; Papadaki, Maria; Al Barghuthi, NedaaBaker(2018) Design of the Blockchain Smart Contract: A Use Case for Real Estate, Journal of Information Security, 9, 177-190.

Kaulartz, Markus; Heckmann, Jörn(2016), Smart Contracts – Anwendungen der Blockchain-Technologie, Computer Und Recht, 32(9).

Lauslahti, Kristian; Mattila, Juri; Seppälä, Timo(2017), Smart Contracts – How will Blockchain Technology Affect Contractual Practices?, ETLA Reports No 68.

Levi,Stuart D.; Lipton, Alex B.(2018), An Introduction to Smart Contracts and Their Potential and Inherent Limitations, Harvard Law School Forum on Corporate Governance and Financial Regulation, available at https://corpgov.law.harvard.edu/2018/05/26/an-introduction-to-smart-contracts-and-their-potential-and-inherent-limitations/#15 (last access: March 25, 2019).

Levy, KarenE. C. (2017), Book-Smart, Not Street-Smart: Blockchain-Based Smart Contracts and The Social Workings of Law, Engaging Science, Technology, and Society 3 (2017), 1-15.

Mik, Eliza(2017), Smart Contracts: Terminology, Technical Limitations and Real World Complexity, Law, Innovation & Technology 9.2.

Nakamoto, Satoshi(2008), Bitcoin: A peer-to-peer electronic cash system 1, White Paper.

Peters, Gareth; Panayi, Efstathios (2016),Understanding Modern Banking Ledgers Through Blockchain Technologies: Future of Transaction Processing and Smart Contracts on the Internet of Money, Banking Beyond Banks and Money: A Guide to Banking Services in the Twenty-First Century, New Economic Windows, Springer.

Posner, Richard A.(1998), Economic Analysis of Law, Aspen Law & Business.

Posner, Richard A. (2004), The Law and Economics of Contract Interpretation, 83 Texas Law Review 1581.

Raskin, Max(2017), The Law and Legality of Smart Contracts, Georgetown Law Technology Review 305.

Savelyev, Alexander(2017), Contract law 2.0: “Smart” contracts as the beginning of the end of classic contract law, Information & Communications Technology Law, 26(2), 116–134.

Shavell, Steven(2007), Contractual Holdup and Legal Intervention, The Journal of Legal Studies Vol. 36, No. 2 (June 2007), pp. 325-354.

Shavell, Steven(2003), Economic Analysis of Contract Law, Discussion Paper No. 403 02/2003, available at http://www.law.harvard.edu/programs/olin_center/papers/pdf/403.pdf ((last access: July 20, 2019).

Shavell, Steven(2005), Is Breach of Contract Immoral?, Harvard Law and Economics Discussion Paper No. 531.

Siegel, David(2016), Understanding The DAO Attack, www.coindesk.com, available https://www.coindesk.com/understanding-dao-hack-journalists.

Sklaroff, Jeremy M.(2017), Smart Contracts and the Cost of Inflexibility, 166 U. Pa. L. Rev. 263.

Tapscott, Dan; Tapscott, Alex(2016), The Impact of the Blockchain Goes Beyond Financial Services, Harvard Business Review, available at https://hbr.org/2016/05/theimpact-of-the-blockchain-goes-beyond-financial-services.

Vatiero, Massimiliano(2018), Smart contracts and transaction costs, Discussion Papers del Dipartimento di Economia e Management – Università di Pisa, n. 238.

Voshmgir, Shermin(2019), Token Economy, O’Reilly.

Werbach, Kevin; Cornell, Nicolas(2017), Contracts Ex Machina, 67 Duke Law Journal 313.

Wright, Aaron; De Filippi, Primavera (2015),Decentralized Blockchain Technology and the Rise of Lex Cryptographia, Social Science Research Network.

Statutory Declaration

I hereby declare and confirm that this thesis is entirely the result of my own work except where otherwise indicated. I acknowledge the supervision and guidance I have received from Professor Carlos Gomez Ligüerre. This thesis is not used as part of any other examination and has not yet been published.

22.07.2019 Felix Schiller

Abbildung in dieser Leseprobe nicht enthalten

Date Name Signature

[...]

---

¹See Dan Tapscott & Alex Tapscott,The Impact of the Blockchain Goes Beyond Financial Services, Harvard Business Review (2016), available at https://hbr.org/2016/05/the-impact-of-the-blockchain-goes-beyond-financial-services (accessed 1 May 2019).

²See Satoshi Nakamoto,Bitcoin: A peer-to-peer electronic cash system 1, White Paper (2008).

³See Mark Giancaspro,Is a ‘smart contract’ really a smart idea? Insights from a legal perspective, Computer Law & Security Review 33 825–835 (2017).

⁴See Ioannis Karamitsos et al.,Design of the Blockchain Smart Contract: A Use Case for Real Estate,Journal of Information Security, 9, 177-190 (2018).

⁵See Benito Arruñada, Blockchain’s Struggle to Deliver Impersonal Exchange, Minnesota Journal of Law, Science & Technology (2018).

⁶See Massimiliano Vatiero,Smart contracts and transaction costs, Discussion Paper n. 238 (2018).

⁷See Guido Governatori et al.,On legal contracts, imperative and declarative smart contracts, and blockchain systems, Artif. Intell. Law 26: 377 (2018).

⁸See Benito Arruñada,Blockchain’s Struggle to Deliver Impersonal Exchange, Minnesota Journal of Law, Science & Technology (2018).

⁹See Ioannis Karamitsos et al.,Design of the Blockchain Smart Contract: A Use Case for Real Estate,Journal of Information Security, 9, 177-190 (2018).

¹⁰See Massimiliano Vatiero,Smart contracts and transaction costs, Discussion Paper n. 238 (2018).

¹¹See Vitalik Buterin,A next-generation smart contract and decentralized application platform,White Paper (2013).

¹²See Benito Arruñada,Blockchain’s Struggle to Deliver Impersonal Exchange, Minnesota Journal of Law, Science & Technology (2018).

¹³See Aaron Wright & Primavera De Filippi,Decentralized Blockchain Technology and the Rise of Lex Cryptographia, Social Science Research Network (2015).

¹⁴See Massimiliano Vatiero,Smart contracts and transaction costs, Discussion Paper n. 238 (2018).

¹⁵See Aaron Wright & Primavera De Filippi,Decentralized Blockchain Technology and the Rise of Lex Cryptographia,Social Science Research Network (2015).

¹⁶See Karen E. C. Levy,Book-Smart, Not Street-Smart: Blockchain-Based Smart Contracts and The Social Workings of Law, Engaging Science, Technology, and Society 3, 1-15 (2017).

¹⁷See Livio Bartoletti & Massimo Pompianu,An empirical analysis of smart contracts: platforms, applications, and design patterns, Lecture Notes in Computer Science (2017).

¹⁸See Lauslahti et al.,Smart Contracts – How will Blockchain Technology Affect Contractual Practices?, ETLA Reports No 68 (2017).

¹⁹See Gareth Peters & Efstathios Panayi,Understanding Modern Banking Ledgers Through Blockchain Technologies: Future of Transaction Processing and Smart Contracts on the Internet of Money, Banking Beyond Banks and Money: A Guide to Banking Services in the Twenty-First Century (pp. 239-279), New Economic Windows, Springer (2016).

²⁰See Livio Bartoletti & Massimo Pompianu,An empirical analysis of smart contracts: platforms, applications, and design patterns, Lecture Notes in Computer Science (2017).

²¹See Livio Bartoletti & Massimo Pompianu,An empirical analysis of smart contracts: platforms, applications, and design patterns, Lecture Notes in Computer Science (2017).

²²See Jenny Cieplak & Simon Leefatt,Smart Contracts: A Smart Way to Automate Performance, Geo. L. Tech. Rev. 417 (2017).

²³See Shermin Voshmgir,Token Economy(O’Reilly, 2019).

²⁴See David Siegel, Understanding The DAO Attack, www.coindesk.com, available at https://www.coindesk.com/understanding-dao-hack-journalists (accessed 22 July 2019).

²⁵See Jenny Cieplak & Simon Leefatt,Smart Contracts: A Smart Way to Automate Performance, Geo. L. Tech. Rev. 417 (2017).

²⁶See Benito Arruñada,Blockchain’s Struggle to Deliver Impersonal Exchange, Minnesota Journal of Law, Science & Technology (2018).

²⁷See Jenny Cieplak & Simon Leefatt,Smart Contracts: A Smart Way to Automate Performance, Geo. L. Tech. Rev. 417 (2017).

²⁸See Stuart D. Levi & Alex B. Lipton,An Introduction to Smart Contracts and Their Potential and Inherent Limitations, Harvard Law School Forum on Corporate Governance and Financial Regulation (2018).

²⁹See Stuart D. Levi & Alex B. Lipton,An Introduction to Smart Contracts and Their Potential and Inherent Limitations, Harvard Law School Forum on Corporate Governance and Financial Regulation (2018).

³⁰See Alexander Savelyev,Contract law 2.0: “Smart” contracts as the beginning of the end of classic contract law, Information & Communications Technology Law, 26(2), 116–134 (2017).

³¹See Jenny Cieplak & Simon Leefatt,Smart Contracts: A Smart Way to Automate Performance, Geo. L. Tech. Rev. 417 (2017).

³²See Pierluigi Cuccuru, Beyond bitcoin: an early overview on smart contracts, International Journal of Law and Information Technology, 25(3), 179–195 (2017).

³³See Chirag Bhardwaj,What Are Smart Contracts: Advantages, Limitations, and Use Cases, available at https://appinventiv.com/blog/smart-contract-guide, last checked on the 14.04.2019.

³⁴See Valentina Gatteschi et al.,Blockchain and Smart Contracts for Insurance: Is the Technology Mature Enough?, Future Internet 10, no. 2: 20 (2018).

³⁵See Chirag Bhardwaj,What Are Smart Contracts: Advantages, Limitations, and Use Cases, available at https://appinventiv.com/blog/smart-contract-guide, last checked on the 14.04.2019.

³⁶See Max Raskin,The Law and Legality of Smart Contracts, Geo. L. Tech. Rev. 305 (2017).

³⁷See Karen E. C. Levy, Book-Smart, Not Street-Smart: Blockchain-Based Smart Contracts and The Social Workings of Law, Engaging Science, Technology, and Society 3, 1-15 (2017).

³⁸See Joshua Gans,The Fine Print in Smart Contracts, NBER Working Paper No. 25443 (2019).

³⁹See Steven Shavell,Economic Analysis of Contract Law, Discussion Paper No. 403 (2003).

⁴⁰See Kristian Lauslahti et al.,Smart Contracts – How will Blockchain Technology Affect Contractual Practices?, ETLA Reports No 68 (2017).

⁴¹See Kevin Werbach & Nicolas Cornell,Contracts Ex Machina, 67 Duke Law Journal 313 (2017).

⁴²See Kevin Werbach & Nicolas Cornell,Contracts Ex Machina, 67 Duke Law Journal 313 (2017).

⁴³See Markus Kaulartz & Jörn Heckmann,Smart Contracts – Anwendungen der Blockchain-Technologie, Computer Und Recht, 32(9) (2016).

⁴⁴See Stuart D. Levi & Alex B. Lipton,An Introduction to Smart Contracts and Their Potential and Inherent Limitations, Harvard Law School Forum on Corporate Governance and Financial Regulation (2018).

⁴⁵See Vitalik Buterin,A next-generation smart contract and decentralized application platform 14, White Paper (2013).

⁴⁶See Valentina Gatteschi et al.,Blockchain and Smart Contracts for Insurance: Is the Technology Mature Enough?, Future Internet 10, no. 2: 20 (2018).

⁴⁷See Pierluigi Cuccuru,Beyond bitcoin: an early overview on smart contracts, International Journal of Law and Information Technology, 25(3), 179–195 (2017).

⁴⁸See Pierluigi Cuccuru,Beyond bitcoin: an early overview on smart contracts, International Journal of Law and Information Technology, 25(3), 179–195 (2017).

⁴⁹See Imran Bashir,Mastering Blockchain(Packt Publishing Ltd., 2017), Chapter 6.

⁵⁰See Imran Bashir,Mastering Blockchain(Packt Publishing Ltd., 2017), Chapter 6.

⁵¹See Guido Governatori et al.,On legal contracts, imperative and declarative smart contracts, and blockchain systems, Artif. Intell. Law 26: 377 (2018).

⁵²See Mark Giancaspro, Is a ‘smart contract’ really a smart idea? Insights from a legal perspective, Computer Law & Security Review 33 825–835 (2017).

⁵³See Guido Governatori et al.,On legal contracts, imperative and declarative smart contracts, and blockchain systems, Artif. Intell. Law 26: 377 (2018).

⁵⁴See Sinclair Davidson et al.,Economics of Blockchain, Public Choice Conference, Fort Lauderdale, United States (2016).

⁵⁵See Karen E. C. Levy,Book-Smart, Not Street-Smart: Blockchain-Based Smart Contracts and The Social Workings of Law, Engaging Science, Technology, and Society 3, 1-15 (2017).

⁵⁶See Karen E. C. Levy,Book-Smart, Not Street-Smart: Blockchain-Based Smart Contracts and The Social Workings of Law, Engaging Science, Technology, and Society 3, 1-15 (2017).

⁵⁷See Levy,supranote 55.

⁵⁸See Robert Cooter et al.,Bargaining in the Shadow of the Law: A Testable Model of Strategic Behavior, Journal of Legal Studies 11(2): 225-251 (1982).

⁵⁹See Karen E. C. Levy,Book-Smart, Not Street-Smart: Blockchain-Based Smart Contracts and The Social Workings of Law, Engaging Science, Technology, and Society 3, 1-15 (2017).

⁶⁰See Richard Posner,Economic Analysis of Law(Aspen Law & Business, 1998), 13–14.

⁶¹See Steven Shavell,Is Breach of Contract Immoral?, Harvard Law and Economics Discussion Paper No. 531 (2005).

⁶²See Kevin Werbach & Nicolas Cornell,Contracts Ex Machina, 67 Duke Law Journal 313 (2017).

⁶³See Richard Posner,The Law and Economics of Contract Interpretation, 83 Texas Law Review 1581 (2004).

⁶⁴See Richard Holden & Anup Malani, Can Blockchain Solve the Hold-up Problem in Contracts?, NBER Working Paper No. 25833 (2019).

⁶⁵See Oliver Hart,Incomplete Contracts and Control, American Economic Review 2017, 107(7): 1731–1752 (2017).

⁶⁶See Steven Shavell,Contractual Holdup and Legal Intervention, The Journal of Legal Studies Vol. 36, No. 2 (2007).

⁶⁷See Richard Holden & Anup Malani,Can Blockchain Solve the Hold-up Problem in Contracts?, NBER Working Paper No. 25833 (2019).

⁶⁸See Richard Holden & Anup Malani,Can Blockchain Solve the Hold-up Problem in Contracts?, NBER Working Paper No. 25833 (2019); See also Shavell, supra note 66.

⁶⁹See Karen E. C. Levy,Book-Smart, Not Street-Smart: Blockchain-Based Smart Contracts and The Social Workings of Law, Engaging Science, Technology, and Society 3, 1-15 (2017).

⁷⁰See Karen E. C. Levy,Book-Smart, Not Street-Smart: Blockchain-Based Smart Contracts and The Social Workings of Law, Engaging Science, Technology, and Society 3, 1-15 (2017).

⁷¹See Sinclair Davidson et al.,Economics of Blockchain, Public Choice Conference, Fort Lauderdale, United States (2016).

⁷²See Lin William Cong & Zhiguo He,Blockchain Disruption and Smart Contracts, NBER Working Paper No. 24399 (2018).

⁷³See Valentina Gatteschi et al.,Blockchain and Smart Contracts for Insurance: Is the Technology Mature Enough?, Future Internet 10, no. 2: 20 (2018).

⁷⁴See Lin William Cong & Zhiguo He,Blockchain Disruption and Smart Contracts, NBER Working Paper No. 24399 (2018).

⁷⁵See Lin William Cong & Zhiguo He,Blockchain Disruption and Smart Contracts, NBER Working Paper No. 24399 (2018).

⁷⁶See Eliza Mik,Smart Contracts: Terminology, Technical Limitations and Real World Complexity, Law, Innovation & Technology 9.2 (2017).

⁷⁷See Imran Bashir,Mastering Blockchain(Packt Publishing Ltd., 2017), Chapter 6.

⁷⁸See Eliza Mik,Smart Contracts: Terminology, Technical Limitations and Real World Complexity, Law, Innovation & Technology 9.2 (2017).

⁷⁹See Valentina Gatteschi et al.,Blockchain and Smart Contracts for Insurance: Is the Technology Mature Enough?, Future Internet 10, no. 2: 20 (2018).

⁸⁰See Eliza Mik, Smart Contracts: Terminology, Technical Limitations and Real World Complexity, Law, Innovation & Technology 9.2 (2017).

⁸¹See Valentina Gatteschi et al.,Blockchain and Smart Contracts for Insurance: Is the Technology Mature Enough?, Future Internet 10, no. 2: 20 (2018).

⁸²See Valentina Gatteschi et al.,Blockchain and Smart Contracts for Insurance: Is the Technology Mature Enough?, Future Internet 10, no. 2: 20 (2018).

⁸³See Eliza Mik,Smart Contracts: Terminology, Technical Limitations and Real World Complexity, Law, Innovation & Technology 9.2 (2017).

⁸⁴See Joshua Gans,The Fine Print in Smart Contracts, NBER Working Paper No. 25443 (2019).

⁸⁵See Max Raskin,The Law and Legality of Smart Contracts, Geo. L. Tech. Rev. 305 (2017).

⁸⁶See Joshua Gans,The Fine Print in Smart Contracts, NBER Working Paper No. 25443 (2019).

⁸⁷See Kevin Werbach & Nicolas Cornell,Contracts Ex Machina, 67 Duke Law Journal 313 (2017).

⁸⁸See Jenny Cieplak & Simon Leefatt,Smart Contracts: A Smart Way to Automate Performance, Geo. L. Tech. Rev. 417 (2017).

⁸⁹See Eliza Mik,Smart Contracts: Terminology, Technical Limitations and Real World Complexity, Law, Innovation & Technology 9.2 (2017).

⁹⁰See Guido Governatori et al.,On legal contracts, imperative and declarative smart contracts, and blockchain systems, Artif. Intell. Law 26: 377 (2018).

⁹¹See Kevin Werbach & Nicolas Cornell,Contracts Ex Machina, 67 Duke Law Journal 313 (2017).

⁹²See Eliza Mik,Smart Contracts: Terminology, Technical Limitations and Real World Complexity, Law, Innovation & Technology 9.2 (2017).

⁹³See Markus Kaulartz & Jörn Heckmann,Smart Contracts – Anwendungen der Blockchain-Technologie, Computer Und Recht, 32(9) (2016).