Blockchain technologies are spreading, attracting both innovative adherents and inflated expectations. Although widely understood as a means of implementing payments in the ‘zero-trust’ spaces of global marketplaces, they have wider implications. This post will consider the potential of robust, secure, distributed and transparent ledgers to support new forms of enterprise, primarily because the technology can (in principle facilitate the productive coordination of individuals and (especially small) enterprises. It does so by allowing individuals or enterprises that are not locked into long-term formal arrangements or relationships to (depending on your perspective) trust each other or to dispense with the need for costly third-party trust verification. In order to realise this potential, however, a range of issues must be tackled, including problems of business culture, the perceptions and processes of regulators and spillovers from the current hype (and anti-hype) surrounding the current headline use of Blockchains to facilitate electronic payments with a high degree of anonymity.
A ledger is a system for recording and transferring value – it thus helps to keep track of the ‘bottom-line’ performance of even a complex entity and to provide a channel for the exchanges among parties who make different contributions to generating and capturing value, with costs and benefits a different times. Therefore, a ledger is a means of compensating costly activity and a way of giving the parties involved a common and (hopefully) reliable insight into how things are going. It is also a special case of a public database, and thus an institutional technology.
A Blockchain is a decentralised ledger platform (Evans 2014). The decentralisation is backed by cryptographic methods which in effect ensure that anyone (or anyone with permission) can see what is written in the ledger, that anyone (with permission from everyone) can modify it, and that anyone using it can be sure that the contents are accurate and authentic. In effect, it keeps a cumulative record of the transactions affecting a given token of value. These are combined cryptographically and new transactions added to the chain through computationally complex activities that a) ensure that the publicly accessible ledger is reliable, up-to-date and visible to all and b) that the process of modification is sufficiently cumbersome tat multiple new transactions involving the same value cannot occur simultaneously (the double-spending problem). These characteristics are obviously valuable in organising private commercial (and other) activity where dedicated rules of access and modification and the need for assured authenticity are of crucial importance. This is not limited to market-facing activities, but extends to dealings between governments and businesses (as we will discuss below). Payments obviously require these characteristics; they must be legal tender in the sense that people receiving a claim on value must know that it can be re-used and will be accepted by others, no matter where it comes from or where it goes to. They must also know (as a matter of common knowledge) that the value involved will be properly transferred in the sense that the recipient fully owns it and the payer no longer owns it. For money transfers, this means that the same token cannot be re-used or re-spent.
This is particularly important in the current Internet-mediated and globalised context, where transfers may need to be: highly automated; too fast for human scrutiny or control; carried out by algorithms; complex in structure; and spread across entities who do not know each other ex ante and have no cost-effective means of verifying identities or following-up agreements. Money itself and other financial instruments deliver some variant of these functions, backed by third parties such as governments or banks. But each of these trust service providers has, for one reason or another, found it necessary or advantageous to limit the scope and speed of value transfer services, giving rise to e.g. exchange rates, strategic transactions costs and potentially fraudulent opportunities for derivative creation and manipulation.
Such manipulation is particularly troublesome for small and innovative enterprises, which typically lack the resources fully to underwrite highly risky ventures and may thus be starved of capital. They cannot easily overcome information asymmetries, and in any case the risks affecting their profitability – and thus the capacity of the economy to identify and direct capital to ‘good idea’ and away from ‘bad ideas’ – depend on a wide range of factors including systemic risk in the economy (which will affect different businesses in different ways) and the incentives driving the behaviour of the firm itself and those to whom it is connected by financial or value chain relationships.
An example will illustrate the point. When a firm borrows seed capital from an investor, it should pay a risk premium to compensate the investor for bearing the risk of failure. At the same time, the premium cannot fully compensate the investor. To agree a contract, therefore, the parties have to come to a common view of the magnitude of the risk and its correlation with other risks (e.g. in the investor’s portfolio). But the relationship does not stop there – if, for example, the investor also funds a rival, demands accelerated payment or calls in the loan or withholds additional financing needed to cope with e.g. delay, the investor can manipulate the probability of success. This is unproblematic providing the investor and the borrower divide the proceeds – both want the venture to succeed. But if the investor can hedge against failure by buying a swap (a CDS) that pays him off in the event of default, this connection is weakened. If (as is unfortunately the case) the investor, or third parties in a position to influence the success of the business can buy an unbounded number of these swaps, they may find themselves actively preferring failure of the original venture. They may, indeed, even seek to encourage ‘start-ups’ that are (or can be made) doomed to fail. Due diligence and transparency should help, but the problem may be complicated by secrecy rules, complexity and changes after the loan has been made and the venture undertaken. The ability to keep track of such related investments and actions in an open, distributed and trustworthy fashion could – in principle - greatly reduce the scope of this problem.
This is not the only solution, of course. As asset pricing models have demonstrated, markets have ways of identifying and pricing (assessing the combined magnitude and severity of) risks. In an ideal world, therefore, the prices of such swaps should provide an accurate measure of risk that does not depend on detailed public accounting and complex modelling of interactions. However, such ‘low-information’ approaches have proven disastrous precisely because of their low-tech nature; informed and uninformed investors can use them, and there is little incentive to work out what the price signals actually mean, when they can rapidly be sold on to other investors guided by the same naïve acceptance of market data. This was one of the main problems associated with the use of the Gaussian Copula formula (Salmon 2012).
Therefore, a suitable stricture of shared accounting for value that could be audited and verified would seem ideal, and Blockchains can certainly perform this function.
But there is nothing in the Blockchain story that requires the ledger system to be used for money transfers. As the Economist noted in a 2015 article,
“Ledgers that no longer need to be maintained by a company—or a government—may in time spur new changes in how companies and governments work, in what is expected of them and in what can be done without them.”
The growth of Blockchains has been associated to economies of scale and scope driven by Moore’s, Kryder’s and Nielsen’s Laws (see Davidson et. al. 2016). In the context of small and agile enterprises, these advantages are magnified when the underlying technology is applied first to more complex transactions than payments and second, to the structure of industry itself. This can be understood from the perspective of Transactions Cost Economics (Williamson 1979); Blockchains are a more cost-effective way of carrying out transactions, minimising not only the data-handling costs but also the agency costs associated with e.g. monitoring and verification.
What can be accomplished using Blockchains instead of firms or markets? Following Williamson, inefficiency in governance (and thus the blocking of some innovations) arises from the conjunction of bounded rationality (and incomplete information with asset specificity and opportunism. Opportunism is generally controlled through markets or hierarchical organisations, which impose hefty costs in exchange, including an embedded resistance to even the most productive disruptive change (though hierarchies are somewhat better than markets). But Blockchains tackle opportunism by technical means and economic incentives, effectively creating spot markets to carry forward what is in essence a pure promise of behaviour to create a durable asset without individual commitment.
But hierarchies and relational contracts did not come into existence only to tackle opportunism. Hierarchies hierarchies exploits the incompleteness of contracts (using ‘position’ as a token) and relational contracting requires and rewards trust among parties. But Blockchains require contractual completeness while firms constitute networks of incomplete contracts.
It seems possible, in theory at least, to replace a complete network of incomplete contracts with an incomplete network of complete contracts. Put slightly differently, the ‘anonymity’ of a Blockchain is qualified – the identities of payers and receivers are securely unknown, but the path taken by tokens of value (the transactions themselves) are known with certainty. What matters is this; how much information must be known, and to whom, in order for the system to function?
Money itself has followed a trajectory that illustrates the point. Originally, cash thrived because it was legal tender, but also because it was anonymous as to both the identities of the parties and the previous history (except for being anchored to a root authority who struck the coins). It created opportunities for strategic corruption (including forgery and theft) and was costly to protect, transport and move over places and times. Therefore, alternatives (financial assets, credit cards, etc.) were developed that dealt with the new problems – but in the process they created ‘traces’ that weakened anonymity (for good or ill). Bitcoins and other cryptocurrencies based on Blockchain technology began as a way to restore what was lost without losing the transactional and ICT-friendly advantages of other virtual payment schemes.
The same applies to contracts. Instead of human-executed contracts that explicitly and permanently define their terms, we can now use smart contracts (self-executing digital contracts - Szabo 2005) to govern interactions and transactions where humans cannot (for reasons of scale, complexity and velocity) venture and thus to connect organisations in a more agile, intelligent and innovation-friendly fashion. This, in turn, leads to a host of effectively decentralised applications including Distributed Autonomous Organisations (DAOs – Mainelli and Smith 2015). These connections already dominate thinking about the architecture and governance of the internet of things (IoT), which ultimately requires decentralised registers for reasons of scale.
The IoT example also points up a further possibility, that Blockchains might use limited memory. At the moment, sufficiently extended Blockchains become ever more cumbersome to use and slow down. They also suffer from the problems that refreshment cycles become long (a matter of minutes) relative to the speed with which the value can be ‘spent’ and that compromises in the record that necessitate killing a specific Blockchain – a principle trust mechanism - can effectively result in ex post repudiation of other transactions along the chain. But these can be mitigated by a ‘moving window’ whereby early parts of the history can be separated or replaced with ‘catch-up’ links, creating an authentic record while preserving practicality.
 Examples are provided by e.g. collateralised debt obligations where a host of assets of different degrees of risk can be aggregated and re-divided into tranches ranging from the secure and expensive to the highly risky and cheap. This recombination, and the ability separately to contract for the underlying debts (through credit default swaps) makes it almost impossible to assess the true value and price of risk or even to track the flow of value through the system.
 In other words, the value of an investment to a risk-averse investor does not depend solely on the probability of profit or loss, but on the correlation of these events with profits or losses in his other investments. If the risks of the investment are negatively correlated with other assets, they help to smooth the returns to the portfolio and generate additional benefits for the investor; if they are uncorrelated or positively correlated, they will leave risk the same or even magnify it. This cannot be determined simply from the expected return on the investment. If the ‘portfolio’ is taken to be the whole market, this incremental contribution to performance is the ‘beta’ in the Capital Asset Pricing Model (CAPM).
 Salmon, Felix. "The formula that killed Wall Street." Significance 9, no. 1 (2012): 16-20.
 Processing costs halve every 18 months.
 Storage costs halve every 12 months.
 Bandwidth costs halve every 24 months.
 Davidson, Sinclair, Primavera De Filippi, and Jason Potts. "Economics of blockchain." Available at SSRN 2744751 (2016).
 Williamson, Oliver E. "Transaction-cost economics: the governance of contractual relations." The journal of law & economics 22, no. 2 (1979): 233-261.
 Mainelli, M., and Mike Smith. "Sharing ledgers for sharing economies: an exploration of mutual distributed ledgers (aka blockchain technology)." The Journal of Financial Perspectives 3, no. 3 (2015): 38-69.