Blockchain was launched amid the global financial crisis with the goal of decentralising money and payments, to avoid the problems associated with their delegation to central authorities. The fifth LTI Report surveys and discusses theories and evidence to shed light on whether this goal has been reached, or whether it is even possible to reach it.
The promise of blockchain, cryptocurrencies, smart contracts, and decentralised finance is to decentralise money, contracts, and markets. The goal is to use new technology to improve efficiency, while shielding economic agents from the opportunistic behaviour of such large players as financial intermediaries and public institutions. Is it possible to reach that goal, and has it been reached? The fifth LTI Report (Amoussou-Guenou et al. 2026) surveys and discusses theories and evidence shedding light on this question.
Money and payment systems play a key role in our economies. It would be extremely difficult for an economy to operate if agents did not have the ability to use money to pay for goods and services, and to store value. Economists, from Wicksell (1901) to Kiyotaki and Wright (1993), have clarified why money is essential, showing how it facilitates mutually beneficial transactions, by mitigating search costs and market incompleteness.
For money to be useful, it must be possible to verify its ownership. Consider the example of Alice offering money to buy goods or services from Bob. Bob needs to verify that Alice’s offer is trustworthy, i.e. that Alice does have money and the money she has is not fake. In the old days, Alice would show her gold or silver coins and experts would verify their precious metal content. In contrast, now that money is scriptural, to verify the validity of Alice’s offer, one needs to rely on a ledger recording money ownership. Inspecting this ledger, Bob can check that Alice has the money on her account, so that the transaction is valid.
Traditionally, the supply of money and the maintenance of ledgers registering its ownership have been overseen and managed by central authorities, such as central banks and commercial banks. Thus, centralised authorities such as commercial banks and central banks can be thought of as the agents to whom citizens delegate the management of money and payments. 1
For this delegation to be beneficial for citizens, the central authorities must be trustworthy. In contrast, the delegation of money and payment to centralised authorities can be costly for citizens when these centralised authorities cannot be fully trusted, and there is information asymmetry about their actions. 2 For example, Farhi and Tirole (2012) show how banks can take excessive risk and create collective moral hazard, while central banks can be unable to prevent the corresponding destabilisation of the financial and monetary system.
To avoid these problems, it could be useful to decentralise the supply of money and the management of the ledger recording its ownership. Decentralising the management of money and payment means delegating it to several agents instead of a single (or a small number of) central authority(ies). This in the line with von Hayek (1976)’s plea for “the replacement of the government monopoly of money by competition in currency supplied by private issuers”. From Hayek’s point of view, the problem with central authorities is that they have a monopoly. When the task is decentralised to several agents, in contrast, even if the agents cannot be individually trusted, according to Hayek competition will keep them in check.
Extending the scope of feasible institutional arrangements beyond those available at the time of Hayek, recent advances in computer science have created a new tool for the decentralisation of money: blockchain.
Blockchain was launched amid the great financial crisis, in defiance of centralised authorities such as banks, central banks, and governments. In 2009 the first block in the Bitcoin blockchain (the genesis block) included the words “Chancellor on brink of second bailout for banks”, which referred to central banks bailing out distressed banks. So, from the start, the goal of blockchain was to decentralise money and payment, to avoid the problems associated with their delegation to central authorities.
The abstract of the seminal white paper of Nakamoto (2008), which was at the origin of the Bitcoin blockchain, expressed this goal: 3
A purely peer-to-peer version of electronic cash [allowing] online payments to be sent directly from one party to another without going through a financial institution…
A blockchain is a decentralised ledger in which the information is structured in blocks of data, and which is replicated among different participants maintaining it, without not necessarily knowing or trusting each other. With blockchain, the management of the ledger, instead of being in the hands of a single central authority, is performed by a network of nodes. That is why it is decentralised. 4
For the decentralised ledger implemented by the blockchain to be useful, it must be that all recorded transactions are valid. 5 Validity implies, in particular, that, as in the case of Alice and Bob example above, the buyer actually owns the money she promises to transfer to the buyer as payment. 6 Since the nodes participating in the blockchain process are in charge of the verification of the validity of the transactions registered in the decentralised ledger, they are often referred to as ‘validators’.
One of the main challenges with blockchains is to ensure that the validators reach a consensus on the ledger. When there is consensus, all participants have locally the exact same ledger, information added in the ledger of one participant is added in the ledger of the other participants in the same order, and all transactions in the ledger have been verified by validators.
How can individual nodes, who don’t know one another, reach such a consensus? The answer is that the design of a blockchain includes the specification of a protocol, suggesting the sequence of actions that nodes should take to process transactions and maintain the ledger.
Figure 1 The average time between consecutive blocks in Bitcoin is stable at around 10 minutes
But will the nodes follow the protocol? After all, there is no central authority to control them, make sure they take the right action, and punish them if they don’t. In practice, the Nakamoto protocol, used by Bitcoin has operated faultlessly since 2008. There has been no successful attack, and the Bitcoin blockchain has proceeded steadily, adding one block (almost exactly) every 10 minutes as prescribed by the protocol (see Figure 1), without any intervention from a central authority. This impressive success is likely to be due to Nakamoto (2008)’s key insight that the blockchain nodes will follow the protocol if it is in their interest to do so, i.e., if the protocol is incentive compatible. As discussed by Biais et al. (2019a, 2019b) and Amoussou-Guenou et al. (2024), two important instruments are used to provide incentives for blockchain validators.
Thus, blockchain technology has made it possible to decentralise money, by supporting cryptocurrencies whose ownership is registered on blockchain.Pagnotta (2022) offers an insightful early analysis of money decentralisation with blockchain. The two largest cryptocurrencies are bitcoin and ether. At the time of writing (February 2026), the total stock of bitcoins is worth more than $1 trillion (more than 4 % of US GDP) and that of ether almost half a trillion dollars. This is quite an achievement.
Once a decentralised ledger and native cryptocurrencies have been successfully deployed, one can move to the second stage: decentralised contracts (i.e. smart contracts). A smart contract is a computer program, deployed on blockchain, whose goal is to execute prespecified instructions, in a decentralised manner, when prespecified conditions are met (Buterin 2014).
Importantly, a smart contract can be written and deployed by anyone, an individual or a group. Unlike traditional contracts, which presuppose an explicit agreement between identified parties and rely on external enforcement, smart contracts do not require prior consent among specific participants; rather, any agent may choose to interact with them, thereby implicitly accepting the rules encoded in their logic. This shift marks a profound transformation from agreement-based to code-based enforcement of commitments.
Once cryptocurrencies and smart contracts can be registered in a blockchain, it becomes possible to use them to offer financial services. This new development is referred to as ‘decentralised finance’, or DeFi (Niepelt, 2025). The three most frequent types of DeFi smart contracts correspond to lending and borrowing, real world assets tokenisation, and decentralised exchanges. Stablecoins 8 are also an important case of DeFi smart contracts.
Figure 2 Market share of the different builders on the Ethereum blockchain
One of the most promising developments DeFi is decentralised exchanges relying on automated market makers (AMM). AMMs are smart contracts committed to supplying liquidity according to predetermined terms. The literature, however, has shown that market failures characteristic of traditional financial markets, such as adverse selection and front running, also affect AMMs. Moreover, while Defi builds on blockchain whose goal was to avoid intermediaries, it has created new types of intermediaries, such as the “builders”, who set up the blocks of transactions to be appended to the Ethereum blockchain. As shown by Capponi et al. (2024) economies of scale lead to the emergence of large builders, as illustrated in Figure 2, who can extract rents from users and validators.
Blockchain has been extremely successful, but now faces major challenges:
New research and innovative developments are needed to meet these challenges, and we believe they will be particularly fruitful if they combine the expertise of economists with that of computer scientists.
Source : VOXeu
The composition of international reserves is in a constant state of flux. This column identifies…
There are concerns that the widespread adoption of central bank digital currencies could drain bank…
Innovation is widely viewed as the engine of economic growth, but we know surprisingly little…
Land-use regulations, including height limits, affect housing affordability and urban productivity. This column analyses over…
Most debate about AI and jobs still starts with the automation frontier: how many tasks…
The 2024 reform of the EU's Stability and Growth Pact introduced medium-term expenditure paths as…