Dashboards for decentralised finance, such as the first result on Google for “DeFi”, display and rank protocols by a metric called “Total Value Locked”, which is measured as the amount of tokens locked in the smart contracts or blockchain of the protocol in question multiplied by the current exchange rate of the token against a fiat currency, most commonly USD. Protocols usually compete for TVL with some mix of printing tokens in return for locking other tokens in the protocol (increasing the multiplicand of locked token amount) and content marketing and memetic engineering (increasing demand and thus the multiplier of fiat price). Implicit in the metric of TVL (total value locked), the competition to maximise it, and the broad structure of the decentralised finance ecosystem alike is this assumption that value can be measured and homogenised by price. This assumption is false, and its deep permeation is evidence that the web3 ecosystem has been largely coopted by the very incentive structures which it was originally designed to oppose.

There are many bad arguments against decentralised finance. Cases of token shilling or thinly-veneered Ponzi schemes are plentiful, but their existence alone does not invalidate more serious projects. As long as profit incentives are in play, such substanceless copycatting arises in protocol engineering, SaaS startups, and techno clubs alike. Nor is all of DeFi, broadly defined, unhelpful — there has been some genuine innovation in protocol primitives (e.g. AMMs) which could fill useful roles in long-lived economic infrastructure with very low maintenance costs. But the central role of “decentralised finance” is as an organising concept: both as a template for brainstorming, as elements or products present in the traditional finance ecosystem are copied and transmogrified to operate in a distributed context, and a justificatory rationale, as if it were desirable to take whatever it is that “finance” does and “decentralise” it. As both a template for brainstorming and a justificatory rationale, the concept of “decentralised finance” is misguided, and in both cases for the same reason: it takes for granted the structures and practices of finance as existent and postulates that they should be in some sense decentralised, without asking whether these structures are the right ones, whether they serve society at large, or even what their purpose is in the first place. In the absence of deep theoretical and historical engagement, even well-intentioned efforts to “decentralise” are likely to be quickly coopted by existing incentive structures, bases of influence, and ideologies in the service of their own perpetuation, and this is exactly what has happened.

Modern financial systems are sprawling and complicated and extend into many aspects of life. Paying for a sandwich at the corner shop, taking on a loan to attend community college, and issuing shares in a partnership to each of the members all constitute interactions of some sort with a “financial system”, whose boundaries and extent are difficult to precisely ascertain. Financial systems mediate the provision, exchange, and transport of food which allow dense urban populations distant from farms to survive, and the abstractions of deep supply chains which render the purchase of an iPhone in California a means of support for continued oppression in Xinjiang. Financial systems provide loans for the construction of new housing, schools, and theatres, and facilitate competitive incentives whereby companies which reduce their costs by emitting more carbon outcompete ones which do not. Financial systems underpin both the magic of modernity whereby I can video call my parents across an ocean and the tragedy of modernity whereby the same cables over which the bits of my face travel also carry hateful vitriol selected for a social media timeline by a company maximising advertising revenue. To even enumerate the benefits and harms, let alone to disentangle which parts of the system are responsible for which, is a task the daunting scope of which can tempt one to throw up their hands in resignation, but the complexity does not justify pretension. Should one wish to craft an alternative system which shifts these equilibria, one must investigate the underlying legal, cultural, and economic patterns which have led to their emergence and sustenance, and brainstorm changes to fundamental legal, cultural, and economic systems and norms which might foster different equilibria. To act without conducting such an investigation is to throw metal darts in a dark room filled with shifting magnetic fields, nearly all of which vector away from the dartboard.

Interpreting charitably, the ethos of decentralised finance paired with some recognition of these complexities carries as implicit the hypothesis that the ills of modern finance are primarily a result of its being centralised, and that decentralising it will alleviate them. While there may be benefits in certain cases, it is not at all clear that this is the case in general, and in many cases decentralisation may exacerbate the harms. iPhone purchases supporting oppression, competition wrecking the environment, and surveillance capitalism fomenting hatred don’t have much to do with centralisation. Splitting Apple up wouldn’t change the abstraction of supply chains and cost-benefit calculus of legal arbitrage. More producers instead of fewer competing to produce widgets seems likely to make it harder to coordinate or legislate, not easier. More social media companies instead of fewer will just cause the attention bidding wars to escalate. Perhaps new companies would propose alternative platforms with different revenue models, which customers would then instead choose — but the cause of improvement here is then not solely the decentralisation of platforms, but rather such paired with a cultural norm which takes into account the harms of certain structures.

If anything can be said to unite these examples of harm perpetuated by the financial system, it is not centralisation or decentralisation but rather abstraction. Purchasers of iPhones are separated from the workers in Xinjiang whose exploitation they contribute to by many layers of cultural and communication barriers and intricate supply chains which make it difficult to reason about specific contributions to harm and facilitate legal arbitrage. Intermediate producers choosing their suppliers are separated from the impacts of their choices by the abstraction of price, which reduces differences in production practices and environmental sustainability to a single number in a competitive environment where consideration of externalities is selected against. Engineers of machine learning systems which power social media algorithms are separated from the communities they bear some responsibility for tearing apart by the abstractions and metrics of surveys, user statistics, time-on-site, and advertising revenue. If decentralising finance means adding layers of abstraction, quantifying value at each layer, and dialing up the knob of competitive pressure, this merely adds fuel to the fire. Reckoning with the complexities and impact of the financial system and conjecturing real alternatives must start with the recognition that these abstractions have diverged. Price is not value. Taking the blind abstractions of capital flow, decentralising them in some fashion, and optimising “total value locked” is not merely a step in the wrong direction. It is a step in the same direction, that of abstraction, metrification, and optimisation. It is not a change at all.