🔬 physics

Unequal Access to Power in Corruption Networks: Evidence from Colombia

This study analyzes corruption networks in Colombia's territorial press to demonstrate that while women are not entirely absent, they face unequal access to power structures that results in their underrepresentation and less recurrent participation compared to men, highlighting gendered patterns of exclusion rather than a lack of female corruption.

Original authors: Giovanna Rodríguez-García, María Elizabeth Mesa-Pineda, José R. Nicolás-Carlock

Published 2026-07-02

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Giovanna Rodríguez-García, María Elizabeth Mesa-Pineda, José R. Nicolás-Carlock

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 corruption not as a lone wolf stealing from a bank, but as a massive, complex dinner party where the guests trade favors, secrets, and protection. This paper investigates who gets invited to this party in Colombia, who sits at the head of the table, and who is stuck in the corner.

The researchers, Giovanna Rodríguez-García, María Elizabeth Mesa-Pineda, and José R. Nicolás-Carlock, wanted to answer a specific question: Is the reason women appear less often in corruption scandals because they are naturally more honest, or is it because they simply don't get invited to the same rooms where the deals are made?

Here is the breakdown of their findings using simple analogies:

1. The "Old Boys' Club" Dinner (Composition)

The researchers looked at a massive list of corruption cases reported in the Colombian press between 1995 and 2022. They treated every person involved as a guest at a dinner.

The Finding: The guest list is overwhelmingly male. About 82% of the guests are men, and only 18% are women.
The Metaphor: It's like walking into a club where the bouncer (the system of power) has historically only let men in. The women aren't necessarily refusing to enter; the doors to the "rooms where deals happen" (political offices, bureaucratic discretion, informal networks) have been much harder for them to open.
The Ties: It's not just about who is there, but who talks to whom. 70% of the conversations (connections) are between two men. Only 4% are between two women. The network is a "masculinized" web.

2. The VIP Section vs. The Periphery (Position)

In any social network, some people are the "connectors"—the ones who know everyone, link different groups, and show up to every party. These are the people in the "core" of the network.

The Finding: Men dominate the "VIP section." They are more likely to be the ones who show up to multiple different corruption scandals (the researchers call these "recidivists").
The Twist: However, the paper found something surprising. When women do manage to get into the main group (the largest connected cluster), they aren't always stuck in the corner. Some women who appear in multiple cases are actually quite well-connected, sometimes even more so than the men in that specific group.
The Metaphor: It's not that women are completely banned from the dance floor. It's that fewer of them get on the floor in the first place. But once they are there, some of them are dancing right in the middle of the crowd, not just on the sidelines. The exclusion is uneven, not absolute.

3. The "Seat at the Table" (Access to Resources)

This is the most critical part of the study. The researchers looked at what kind of seats men and women were sitting in.

The Finding: Even when women are present, they are more likely to be in seats with less power. They hold roles that have less control over money, less ability to make decisions, and less access to "coercive" power (like police or military influence). Men are much more likely to hold the "power seats"—the high-ranking jobs where the big decisions and big bribes happen.
The Metaphor: Imagine the corruption network as a high-stakes poker game. Men are mostly sitting at the high-stakes tables with the big chips, making the rules. Women are more likely to be at the low-stakes tables or working as the dealers. They are part of the game, but they aren't the ones holding the biggest stack of chips or controlling the flow of the money.

The Big Takeaway

The paper argues that we need to stop asking, "Are women just naturally more honest than men?"

Instead, we should ask, "How does the structure of power keep women out of the rooms where corruption happens?"

The study suggests that corruption is a networked practice. It requires trust, protection, and access to resources. Because these networks have historically been built by and for men (the "Old Boys' Club"), women are structurally excluded from the most powerful spots. When they do get in, it's often through specific, lower-power doors.

In short: The data shows that women aren't necessarily "less corrupt" by nature; they are less incorporated into the specific power structures where corruption is organized. The system doesn't just filter them out; it filters them into different, less powerful roles within the system.

Technical Summary: Unequal Access to Power in Corruption Networks: Evidence from Colombia

Problem Statement
Corruption is fundamentally a networked practice embedded in relational structures that facilitate access to information, resources, coordination, and protection. While extensive literature has debated whether women are inherently less corrupt than men—attributing differences to honesty, risk aversion, or prosociality—this paper argues that such individual-level explanations overlook the structural mechanisms of power. The central problem addressed is whether gendered inequalities in access to political and administrative power are reproduced within corruption networks themselves. Specifically, the authors investigate if women's lower presence in corruption cases reflects a structural exclusion from the informal spaces where corrupt exchanges are organized, rather than an intrinsic lack of corrupt intent.

Methodology
The study utilizes a network science approach applied to empirical data from Colombia between 1995 and 2022.

Data Source: The analysis draws from Transparencia por Colombia's Radiografía de Hechos de Corrupción, a dataset systematizing corruption cases based on territorial press reports and public sources. The final analytical sample comprises 1,080 corruption cases involving 4,168 unique individual actors (excluding institutional or collective actors).
Network Construction: The authors employ a standard case-actor bipartite network approach. They first construct a bipartite graph linking actors to specific corruption cases. This is then projected onto an actor-actor co-participation network, where an edge exists between two actors if they appear in the same corruption case.
Variables and Measures:
- Gender: Actors are classified as Male (M) or Female (F) based on names, acknowledging this as a binary inference rather than self-identified gender.
- Network Position: The study analyzes degree centrality (number of ties), recurrence (actors appearing in multiple cases, termed "recidivists"), and k-core decomposition to identify the most densely connected layers of the network.
- Resource Access: A five-point weighted scale is constructed based on the actor's institutional category and reported position (e.g., elected official vs. administrative staff) to measure access to political, administrative, economic, or coercive resources.
Analytical Strategy: The authors test three hypotheses regarding gendered composition, network position, and institutional segmentation using descriptive statistics, network visualization, and statistical tests (Welch's t-test, permutation tests, and Mann–Whitney U tests).

Key Results
The empirical analysis reveals three distinct dimensions of gendered inequality within Colombian corruption networks:

Masculinized Composition (Support for H1): The networks are strongly masculinized. Men constitute 82% of all actors and 70% of all ties (male-male), while women represent only 18% of actors and 4% of ties (female-female). This imbalance persists among "recidivists" (actors in multiple cases), where men account for 87% of repeat participants.
Uneven Network Position (Partial Support for H2): Contrary to the expectation that women are entirely peripheral, the results show a nuanced pattern. While women are underrepresented in the network overall, those who do appear are not necessarily confined to the periphery.
- Among actors in the Largest Connected Component (LCC), women who appear in multiple cases show slightly higher average connectivity than their male counterparts in the same group.
- In the innermost network core ( $k=37$ ), women represent 40% of actors, a higher proportion than their overall presence in the network.
- This suggests that while access to the network is restricted, women who penetrate the network can occupy central or dense positions; exclusion is uneven rather than absolute.
Institutional Segmentation (Support for H3): A significant gender gap exists in the types of roles women occupy. Women are disproportionately concentrated in positions with lower access to discretionary resources.
- The average resource-access score is 3.06 for men and 2.85 for women ( $p < 0.001$ ).
- This gap is most pronounced in the public sector and among actors linked to the illegal economy. In the public sector, men are significantly more likely to hold high-access positions (weight 5), whereas women are concentrated in intermediate-access roles (weight 3).
- Higher resource access is more consistently linked to stronger network positions for men than for women.

Key Contributions
The paper makes three primary contributions to the literature:

Theoretical Shift: It moves the debate on gender and corruption away from individual traits (honesty, risk aversion) toward a structural "access-to-power" perspective. It posits that women's lower participation is a result of unequal incorporation into the informal spaces where corruption is organized.
Network Analysis: It introduces a gendered and institutional dimension to the analysis of corruption networks, demonstrating that gender disparities are not merely compositional (numbers of actors) but also positional (centrality, recurrence) and institutional (resource access).
Empirical Evidence: It provides the first detailed evidence from Colombia, a context where territorial politics and public contracting are deeply intertwined with corruption, showing how these networks reproduce gendered patterns of exclusion.

Significance and Claims
The authors claim that their findings shift the analytical focus from asking "Are women less corrupt?" to "How are women and men differently incorporated into corrupt networks?" The study argues that corruption networks are not random collections of actors but organized systems that reproduce existing power structures.

The significance of the findings lies in the demonstration that gendered exclusion is structured but not absolute. While women are systematically underrepresented and concentrated in lower-resource roles, they are not entirely absent from central network positions. This suggests that corruption networks are not impenetrable male fortresses but rather structures where access is stratified. The paper concludes that gendered access to power leaves observable traces in the composition, recurrence, and institutional segmentation of corruption networks, supporting the view that corruption is a networked practice embedded in unequal structures of access.

The authors maintain modesty regarding their claims, noting limitations such as the reliance on press-reported data (which captures visible, not necessarily all, corruption), the binary nature of the gender variable, and the use of resource-access weights as analytical proxies rather than direct measures of power. They suggest future research should explore longitudinal changes, comparative regional patterns, and qualitative mechanisms of female incorporation.

1. The "Old Boys' Club" Dinner (Composition)

2. The VIP Section vs. The Periphery (Position)

3. The "Seat at the Table" (Access to Resources)

The Big Takeaway

Technical Summary: Unequal Access to Power in Corruption Networks: Evidence from Colombia

More like this