💰 quantitative finance

Crypto-Microeconomics: The Distribution of Bitcoin Wealth Among Diverse Economic Agents

This paper introduces a "Crypto-Microeconomic Observability Framework" to reveal that Bitcoin wealth is structurally uneven and highly concentrated among diverse economic agents, particularly Service entities and a small subset of Abuse and Individual accounts, sustaining a persistent "Whale-Effect" despite varying entity counts.

Original authors: Syed Azhar Hussain, Kashif Ahmad, Mubashir Husain Rehmani

Published 2026-07-07

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Syed Azhar Hussain, Kashif Ahmad, Mubashir Husain Rehmani

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the Bitcoin network not as a mysterious digital vault, but as a massive, bustling city where everyone owns a digital wallet. For years, economists have looked at this city from a helicopter, counting the total number of coins and the average price of a house (macroeconomics). They've seen that a few people own a lot of money, but they haven't really looked at who those people are or how different groups of people in the city are doing.

This paper builds a "microscope" to look at the ground level. The authors created a new framework called the Crypto-Microeconomic Observability Framework. Think of it as a high-tech sorting machine that takes the chaotic list of every Bitcoin wallet and organizes them into five distinct neighborhoods based on who owns them:

Service: The big businesses (exchanges, payment processors, mining farms).
Abuse: The criminals (scammers, Ponzi schemes, theft rings).
Malware: The hackers (ransomware gangs, crypto-jackers).
Individuals: Regular people (often identified by their forum usernames).
Benign: Publicly known good actors (government addresses, charities).

The Big Discovery: The "Whale" is Everywhere

The paper uses a concept called the "Whale-Effect." Imagine a small pond where a few giant whales hold 99% of the water, leaving tiny fish with almost nothing. The authors found that this isn't just happening in the whole city; it's happening in every single neighborhood.

Here is what they found when they looked at the data:

The Service Neighborhood is the Giant: This group is the biggest. They hold 75% of all the Bitcoin the researchers could identify. They are the "whales" of the business world.
The Abuse Neighborhood is the Shocking Underdog: This is the most surprising finding. The "Abuse" group (criminals) only makes up about 3.5% of the total number of wallets. However, they control a massive 24% of the total Bitcoin wealth. It's like a small gang of thieves in a city of millions holding a quarter of the city's gold.
The Individual Neighborhood is a Mirage: There are thousands of regular people (Individuals) with wallets, but they hold very little money on average. The wealth is so concentrated that a tiny handful of "super-rich" individuals hold almost everything, while the rest have pennies.

The "Empty Wallet" Reality

The paper also points out a funny but important detail: Most wallets are empty.
Out of over 223,000 labeled wallets they looked at, 96% had zero Bitcoin.
It's like walking into a stadium with 223,000 seats, but only 8,450 people are actually sitting there. The rest are empty chairs. Among those who are sitting, the distribution is incredibly unfair.

How Unequal is It? (The "Gini" Score)

The authors used a score called the Gini coefficient to measure inequality.

0 means everyone has exactly the same amount.
1 means one person has everything, and everyone else has nothing.

The results were shocking. In the Individual neighborhood, the score was 0.9993. That is practically perfect inequality. It means that within the group of "regular people," the wealth is so skewed that it looks like one person owns the whole group's money. The Abuse and Service groups were also nearly at this maximum level of inequality.

Does Time Change Anything?

The researchers looked at the data from 2009 to 2025 (including big events like the collapse of exchanges, ransomware attacks, and new laws). They found that the "Whale-Effect" doesn't go away. Even when big things happen in the news, the wealth stays stuck in the hands of a very small number of entities in each category. The structure of the city remains the same: a few giants hold the water, and the tiny fish get the drips.

The Bottom Line

The paper concludes that Bitcoin is not as "democratic" or "decentralized" as we might hope when you look at the actual owners. While there are many different types of owners (businesses, criminals, regular people), the money is not spread out among them. Instead, it is heavily concentrated in a tiny subset of "whales" within every single category.

In simple terms: Bitcoin wealth is like a game where a few players in every category have stacked the deck, and the rest of the players are playing with almost no cards.

Technical Summary: Crypto-Microeconomics and Bitcoin Wealth Distribution

Problem Statement
Existing literature on Bitcoin wealth distribution predominantly relies on macroeconomic indicators, such as aggregate wallet balances, network activity, and top-k wealth shares (e.g., the "Bitcoin Rich List"). While these studies utilize metrics like the Gini coefficient and Nakamoto index to identify broad concentration trends (the "Whale-Effect"), they fail to address microeconomic imbalances among specific market segments. Current approaches often overlook the complex interactions between diverse economic agents—ranging from legitimate service providers to illicit actors—and lack a structured framework to analyze wealth disparity at the individual agent level. Furthermore, the pseudonymous nature of Bitcoin complicates the identification of these agents, leading to a gap in understanding how wealth is distributed across different behavioral categories (e.g., Service, Abuse, Malware, Individuals, and Benign).

Methodology
The authors propose a "Crypto-Microeconomic Observability Framework" designed to systematically assess Bitcoin wealth distribution across five labeled economic agent classes:

Service: Entities handling funds (exchanges, mixers, mining operations, etc.).
Abuse: Entities engaged in illicit activities (Ponzi schemes, theft, terrorism, etc.).
Malware: Operators of malicious software (ransomware, cryptojacking, banking trojans).
Individuals: Users identifiable via forum usernames (e.g., BitcoinTalk).
Benign: Publicly disclosed addresses (government, donations).

The methodology employs a multi-step pipeline:

Data Extraction: A local Bitcoin node extracts the full ledger up to block height 905,314 (July 2025). Data is structured using BitSQL.
Deanonymization and Clustering: Using "Multi-input Classifier" techniques and BlockSci, addresses are clustered into entities and tagged into the five agent categories.
Metric Calculation: The study computes descriptive statistics and specific inequality metrics for entities with positive balances ( $E^+_a$ $E_{a}^{+}$ ):
- Top-K Share Ratios: The percentage of BTC held by the top 1%, 5%, and 20% of positive-balance entities within each class.
- Gini Coefficient: Measuring intra-class wealth inequality.
- Longitudinal Analysis: Time-series tracking of these metrics from 2009 to 2025, overlaid with significant market events (e.g., Halvings, exchange collapses, regulatory actions).

Key Results
The analysis of the dataset (223,558 labeled entities, of which 8,450 hold positive balances) reveals the following:

Structural Concentration: Bitcoin wealth is highly concentrated across all major classes, exhibiting a persistent "Whale-Effect."
- Service Entities: Hold the largest share of observed BTC (75.15%) and constitute 71.98% of the entity count.
- Abuse Entities: Control a disproportionately large share of wealth (24.26% of total BTC) relative to their small entity count (3.53%). They possess the highest mean balance (29.06 BTC) among non-service entities.
- Individuals: While holding only 0.57% of the total supply, this class exhibits near-maximal internal inequality (Gini = 0.9993).
Inequality Metrics:
- Gini Coefficients: Service (0.9935), Abuse (0.9962), and Individuals (0.9993) show near-total wealth concentration within their respective classes. Malware (0.5380) and Benign (0.7239) show lower, though still significant, inequality.
- Top-K Shares: In the Service, Abuse, and Individual classes, the top 1% of entities hold over 95% of the class's total BTC. For Abuse, the Top-1% holds 99.55% of the class's wealth.
Distribution Characteristics: Balances are heavily right-skewed and heavy-tailed. The median balance for most classes is negligible (often near zero), while the mean is driven by a few extreme outliers. Individuals show the highest skewness (62.24) and kurtosis (3,883.87).
Temporal Persistence: Time-series analysis indicates that these concentration patterns are structurally persistent rather than transient. Major market events (e.g., Mt. Gox collapse, FTX failure, Halvings) coincide with shifts in concentration, but the underlying "Whale-Effect" remains dominant across Service, Abuse, and Individual categories.

Significance and Claims
The paper claims to address significant gaps in crypto-economic literature by shifting the focus from macro-level aggregate analysis to a structured microeconomic examination of labeled economic agents. Its primary contributions are:

Framework Development: The introduction of a "Crypto-Microeconomic Observability Framework" that integrates data extraction, deanonymization, and granular wealth analytics.
Agent-Specific Insights: Demonstrating that wealth concentration is not uniform; specific agent types (particularly Abuse and Service) hold disproportionate power relative to their numbers, challenging the assumption that wealth distribution can be understood solely by entity counts.
Micro-Level Whale-Effect: Providing empirical evidence that the "Whale-Effect" operates at the micro-level within specific agent classes, where a tiny subset of entities controls the vast majority of funds within those categories.

The authors conclude that Bitcoin wealth among labeled economic agents remains structurally uneven. They assert that their findings support the need for improved deanonymization techniques to better track micro-level wealth inequality, suggesting that the current distribution mirrors or exacerbates conventional economic disparities rather than democratizing finance. Future work is proposed to extend the framework with enhanced entity-resolution and attribution methods.