To Get Serious About Decentralization, We Need to Measure It

Stephanie Hurder, a CoinDesk columnist, is a founding economist at Prysm Group, an economic advisory focused on the implementation of emerging technologies, and an academic contributor to the World Economic Forum. She has a PhD in Business Economics from Harvard.

On May 15, Decrypt reported that six weeks after being acquired by Binance, the crypto data aggregator CoinMarketCap updated the method by which it ranked exchanges on its site. This update, perhaps unsurprisingly, moved Binance into the top spot. Critics argued CoinMarketCap was basing its rankings on factors that had little to do with the fundamental quality of exchanges, such as liquidity and security. The COO of CoinMarketCap competitor CoinGecko added that CoinMarketCap would need to “go deeper to get a more holistic picture of things.”  

This was not the first data scandal for the popular site, which has committed to refining and expanding its ranking metrics in light of the industry outcry. But it added to a general feeling of frustration regarding the availability of reliable, fair crypto industry metrics. Eustace Cryptus, writing for bitcoinist.com, lamented: “Will CoinMarketCap ever provide accurate crypto data?”  

See also: Stephanie Hurder – How Blockchain Tech Can Make Coronavirus Relief More Effective

Determining a set of metrics that will be used to compare projects, companies or national economies can be a fraught exercise. Even the most stalwart measures of national accounting come with controversy. Gross Domestic Product (GDP) had immediate and vocal opponents after its introduction in 1934. Critics of the unemployment rate argue that excluding anyone who is not actively seeking work from the base population misrepresents the status of the labor market. Yet, both of these measures remain widely applied and newsworthy because they provide insight into important dimensions of the health of the economy.

Despite inevitable controversies, it is essential not to abandon the endeavor of measuring economic systems. It would be far more challenging to grasp the novel coronavirus’s unprecedented impact on the U.S. labor force, for example, without weekly unemployment reports from the Department of Labor.

This is true for blockchain as much as it is for state and national economies. Blockchain-based systems are economies written in code. The value of a protocol, and any native tokens, depend on economic fundamentals. As such, users, validators and investors benefit from a standard set of measures that gauge these fundamentals and evaluate whether systems are working as intended.  

Consider decentralization. Most protocols list decentralization in mining to be a core goal of their project.  But pressed to define what this would be more rigorously, they cannot. Some aim for a minimum number of participating miners – say, 100 – while others suggest gauging decentralization indirectly through profitability or governance. Without an agreed-upon framework, the industry talks past itself.

Economics can provide the framework for the rigorous measurement of outcomes such as decentralization. Measures such as the Herfindahl-Hirschman Index are widely used in the study of industries to capture the distribution of market power of participants. Applied to block production, these metrics gauge to what extent miner influence and rewards are decentralized in practice.

The OAN, or the Open Application Network, used these frameworks to evaluate the impact of a recent protocol change on mining decentralization. Originally a Proof-of-Work protocol, the founding team was concerned a small set of mining pools had an outsized influence on block production. They launched a combined Proof of Work-Proof of Stake consensus mechanism in November 2019 but until recently had few tools to quantify to what extent their decentralization efforts had succeeded. Applying economic metrics, the OAN team determined the inclusion of Proof-of-Stake reduced block production centralization substantially, transforming them from a highly centralized platform to more in line with industry leaders Ethereum and Bitcoin (see Figure 1).  

Having concrete numbers implies projects can be honest about their degree of success in achieving decentralization. It also means the industry can be frank with its overall state. Ethereum’s level of mining decentralization, while moderate, is still closer to the dynamics of having a handful of major block producers than a vast, highly decentralized network. An industry that espouses a goal of hundreds of miners should be aiming for concrete metrics that confirm the achievement of this goal. For the leading protocols, this is still a ways away. 

The benefits of measurement – and the dose of truth it can bring – apply to more than just mining decentralization. While blockchain projects differ in the design of their tokens, consensus and governance, they share common objectives of creating widely used and valuable tokens, democratizing influence and control, and broadening access to capital and data. It may take several months – or years – before the industry converges on a common quantitative framework for measuring progress toward these goals, but the effort is well worth the investment.