Are DAOs Decentralized? Part Two

Answering this question depends as much on whether or not we can define decentralization as it does on whether or not we can come up with a framework for measuring it.

In Part I, we touched on the diversity of the ways people and DAOs define decentralization and talked about why it would be helpful to measure decentralization. In a nutshell, if we’re talking about whether or not a DAO is more or less decentralized than a comparable organization, it would be helpful to have a way to measure the degree of decentralization in different organizations.

Now, I want to go back and look at where academics have landed in terms of identifying and overcoming the challenges of measuring decentralization. This literature review includes peer-reviewed publications found using Google Scholar. They’re generally organized in chronological order, with the exception of the first article, because it provides a great introduction to the issues around defining and measuring decentralization. I’ve also included several publications that are not peer-reviewed for further reading at the end of this article.

1. Decentralization Dilemma: Measuring the Degree and Evaluating the oOtcomes (2006)

Published in The Indian Journal of Political Science, this paper by Chanchal Kumar Sharma actually argues that the diversity of definitions of decentralization makes it impossible to define and measure, suggesting that it’s more important to focus on context than specific measurements.

Sharma emphasizes the fact, also pointed out by Schneider, that decentralization is multifaceted and the lack of a common definition or understanding makes it difficult to compare existing studies on decentralization in particular locations.

Ultimately, Sharma concludes by arguing for “a strictly contextual yet comprehensive approach,” emphasizing that the more data we’re about to include on decentralization, the more detail we can get into about what decentralization is and means.

2. Decentralization: Conceptualization and Measurement (2003)

Aaron Schneider is a political scientist who published this article in the Fall 2003 edition of Studies in Comparative International Development to try and establish a definition of decentralization and offer a way of measuring decentralization based on that definition. At least in the context of political science, Schneider proposes that decentralization includes three dimensions:

• Fiscal: How much money does the central government provide to entities that aren’t a part of the central government, or to what degree does the central government give entities that aren’t a part of the central government power over its budget?

• Administrative: How much power does the central government allow its subordinates to exercise?

• Political: To what degree does the central government allow its subordinates to interact with the public? In other words, do subordinates get to act like governments themselves, or do they report directly to the central government?

Schneider goes about defining decentralization by saying that “Decentralized systems are those in which central entities play a lesser role in any or all of these dimensions.” He goes on to use data from 1996 for sixty-eight countries to test this definition.

Data on subnational expenditures and revenue is used to determine the fiscal dimension, taxes and transfers as a percentage of subnational grants and revenue are used to determine the administrative dimension, and whether or not municipal or state/provincial governments hold elections is used to determine the political dimension.

He concludes by suggesting that this framework be used to reexamine research on decentralization in specific locations and better understand the causes and effects of decentralization over time.

3. Comparative Federalism and Decentralization: On Meaning and Measurement (2004)

Published in Comparative Politics, this article by Jonathan Rodden dives deep into the problems associated with measuring fiscal, political, and administrative decentralization.

Rodden touches on the difficulties in defining decentralization before pointing out that attempts to measure decentralization using quantitative data often fail to provide an effective basis for comparison because they don’t account for what’s happening in the bigger scheme of things.

As an example, Rodden points out that “...Denmark is the third-most decentralized country in the world according to Table 1…though the central government tightly regulates virtually every aspect of local government finance.”

Whether we look at fiscal, policy, or administrative decentralization, we need to understand the context in which data is being collected because different governments handle similar issues like elections and tax collection in very different ways.

Rodden concludes by suggesting we pay careful attention to the limits of qualitative and quantitative data and continue trying to combine these different types of data to get a better sense of the causes and effects of fiscal, political, and administrative decentralization.

4. The Variety of Decentralization Indexes: A Review of the Literature (2021)

Published in Regional and Federal Studies by Jean-Baptiste Harguindéguy, Alistair Cole, Romain Pasquier, this article focuses specifically on indexes used to rank countries and “sub-state units” by the degree of autonomy they exercise over their territory.

The authors compare ten different variables across twenty-five indexes for measuring decentralization, highlighting two methods in particular that they find to be the most effective. In addition to academic frameworks like those we’re talking about, these methods include indexes from significant groups like the International Monetary Fund and the Organization for Economic Cooperation and Development.

The authors reaffirm the importance of considering the political, administrative, and fiscal dimensions of decentralization. They also confirm Roddin’s association of decentralization with federalism by pointing out that the development of indexes for measuring decentralization stems from political scientists interested in comparative federalism and economists who’ve gathered fiscal data on various governments.

The authors conclude by stating their preference for the indexes from the Regional Authority Index and Local Authority Index, mostly due to the comprehensiveness of their data and transparency in their methodology. Compiling the indexes seems to involve a significant degree of interpretation in terms of what gets defined as being part of the central state versus sub-state units. On one hand, we can’t compare countries if we can’t fit them into the framework. On the other, the way researchers go about defining these terms introduces a lot of room for skepticism toward the results.

5. When is a DAO Decentralized? (2022)

In this article, published by Henrik Axelsen, Johannes Rude Jensen, and Omri Ross in Complex Systems Informatics and Modeling Quarterly, the authors introduce a framework for measuring decentralization in DAOs based on the idea of “sufficient decentralization.”

Sufficient decentralization is defined as “...a verifiable state, where (1) the design of the DAO is collusion resistant and based on long-term equilibrium; (2) its governance processes have unrestricted and transparent access.” In other words, the authors aim to measure decentralization in terms of how well an organization can avoid being damaged by “centralized” competitors.

The framework consists of a questionnaire that aims to assess the level of decentralization in a DAO across five categories of questions:

• Token-weighted voting and incentives
• Infrastructure
• Governance
• Escalation
• Reputation

The authors explain that in this context, decentralization is defined as a response to issues in systems perceived to be “centralized.”

The researchers conducted eight 45-60 minute interviews to draw their conclusions and while it’s become popular to talk about decentralization in terms of an organization’s ability to avoid regulation, the only basis for this definition of “sufficient decentralization” appears to be a couple of blog posts by Ethereum co-inventor Vitalik Buterin. By this definition, even if the decentralized alternative takes the same form as the centralized system, the decentralized alternative is deemed “decentralized” because it rejects any association with the “centralized” system.

While I believe the results of this article and its framework should be taken with a grain of salt, the authors make an important point in mentioning that the development of decentralization in various organizations should be studied in addition to the relative levels of decentralization in different organizations.

These articles touch on a number of important things we should consider when it comes to measuring decentralization in DAOs. Sharma’s suggestion that we need a “contextual yet comprehensive approach” provides a great place to start.

While the framework that’s been developed around political, fiscal, and administrative decentralization gives us a helpful starting point for measuring different types of decentralization, the literature also shows that it can be difficult to compare governments due to the different ways they manage their finances and governance. Similarly, we can expect to run into difficulties measuring decentralization in DAOs because different DAOs have different ways of managing their finances and governance. This underscores the fact that developing this type of framework inherently involves an ongoing conversation about the utility of the framework and what can be done to improve the way we measure decentralization.

To piggyback on the work people have already put into defining decentralization, the definition provided in The Variety of Decentralization Indexes: A Review of the Literature appears to be the most accurate and up-to-date. While the authors acknowledge that academics recognize different definitions of decentralization, they also argue that all indexes of decentralization embrace the idea that measuring decentralization is about getting a sense of the way central governments and “sub-state units” coordinate self-rule and shared rules between themselves.

Elazar (1987) initially defined self-rule and shared-rule as the political, administrative and fiscal powers of sub-state units in their own jurisdictions, along with those they share with central government, respectively. According to this approach, the self-rule/shared-rule distinction allows capturing variations across sub-dimensions of the concept of federalism and doing so in “degrees” on continuous or ordinal scales. (p5)

In other words, measuring decentralization is always about finding ways to measure how much autonomy different groups and/or actors have within organizations and how much autonomy organizations have relative to each other. Political, fiscal, and administrative decentralization represent different ways of measuring this autonomy:

• Political decentralization is about getting a sense of how many levels of hierarchy exist in an organization. More levels provide more opportunities for representation, which translates into a more decentralized organization.

• Fiscal decentralization is about getting a sense of the impact each level has in relation to the overall impact of the organization. If

• Last, but not least, administrative decentralization is about getting a sense of the degree to which each “level” can act on its own authority.

Examining each dimension of decentralization on its own and comparing it with the other dimensions helps paint a more complete picture than if we just considered one of the dimensions on its own.

As I mentioned before, when we talk about how to go about measuring different types of decentralization in the next article, it’s important to keep in mind that any framework we come up with will be inherently incomplete - or at least, inherently tailored to the specific form of decentralization we want to measure. Any framework we start with will become part of an ongoing discourse about measuring decentralization and will need to be modified and expanded as DAOs experiment with new ways of managing their governance and finances.

Still, we’ve got to start somewhere. Now that we’ve reviewed the literature on measuring decentralization, we can start talking about how to measure decentralization in DAOs in Part III.

Critique

1. Sharma, C. K. (2006). Decentralization dilemma: measuring the degree and evaluating the outcomes. The Indian Journal of Political Science, 49-64. https://mpra.ub.uni-muenchen.de/204/1/MPRA_paper_204.pdf

2. Schneider, A. (2003). Decentralization: Conceptualization and measurement. Studies in Comparative International Development, 38(3), 32–56. https://doi.org/10.1007/bf02686198 http://projects.mcrit.com/foresightlibrary/attachments/article/1234/Schneider_Decentralization.pdf

3. Rodden, J. (2004). Comparative federalism and decentralization: On meaning and measurement. Comparative politics, 481-500. http://fdjpkc.fudan.edu.cn/_upload/article/files/b4/7f/51b246194ea680cdb47c1bb020c5/6f59d35d-96fe-47c7-adce-44c93bf61b0c.pdf

4. Jean-Baptiste Harguindéguy, Alistair Cole, Romain Pasquier. The variety of decentralization indexes: a review of the literature. Regional and Federal Studies, 2021, 31 (2), pp.185-208. ff10.1080/13597566.2019.1566126ff. Ffhal-02055944 https://hal-univ-rennes1.archives-ouvertes.fr/hal-02055944/document

5. Axelsen, H., Jensen, J. R., & Ross, O. (2022). When is a DAO Decentralized?. Complex Systems Informatics and Modeling Quarterly, (31), 51-75. https://csimq-journals.rtu.lv/article/viewFile/csimq.2022-31.04/2972

1. Measuring Decentralization and the Local Public Sector: A Survey of Current Methodologies

a. Abdelhak, F., Chung, J., Du, J., & Stevens, V. (n.d.). Measuring Decentralization and the Local Public Sector: A Survey of Current Methodologies. Retrieved January 9, 2023, from https://www.urban.org/sites/default/files/publication/23241/412733-Measuring-Decentralization-and-the-Local-Public-Sector-A-Survey-of-Current-Methodologies.PDF

2. Devometrics: How to measure decentralisation? A review of the literature

a. Dardanelli, P., Wright, K. (2021, July 6). Devometrics: How to measure decentralisation? A review of the literature. LGiU. Retrieved January 8, 2023, from https://lgiu.org/publication/devometrics-how-to-measure-decentralisation-a-review-of-the-literature/

3. Decentralization Measures Revisited

a. Martinez-Vazquez, J., & Timofeev, A. (2010, April). Decentralization Measures Revisited. Retrieved January 8, 2023, from https://www.researchgate.net/profile/Jorge-Martinez-Vazquez/publication/46455685_Decentralization_Measures_Revisited/links/0912f5098cd8ebb329000000/Decentralization-Measures-Revisited.pdf

4. Defining and measuring decentralisation: a critical review

a. Torrisi, G., Pike, A., Tomaney, J., & Tselios, V. (2011). Defining and measuring decentralisation: a critical review. https://www.semanticscholar.org/paper/Defining-and-measuring-decentralisation%3A-a-critical-Torrisi-Pike/a428bdcf65fce1458f73e417155c3583d2b8cbcc