An Algorithmic Introduction to Savings Circles

Rediet Abebe, Adam Eck, Christian Ikeokwu, Samuel Taggart

Abstract

Rotating savings and credit associations (roscas) are informal financial organizations common in settings where communities have reduced access to formal financial institutions. In a rosca, a fixed group of participants regularly contribute sums of money to a pot. This pot is then allocated periodically using lottery, aftermarket, or auction mechanisms. Roscas are empirically well-studied in economics. They are, however, challenging to study theoretically due to their dynamic nature. Typical economic analyses of roscas stop at coarse ordinal welfare comparisons to other credit allocation mechanisms, leaving much of roscas' ubiquity unexplained. In this work, we take an algorithmic perspective on the study of roscas. Building on techniques from the price of anarchy literature, we present worst-case welfare approximation guarantees. We further experimentally compare the welfare of outcomes as key features of the environment vary. These cardinal welfare analyses further rationalize the prevalence of roscas. We conclude by discussing several other promising avenues.

Introduction

Rotating Savings and Credit Associations (ROSCAs) are informal financial institutions prevalent in communities with limited access to formal banking. These groups, known by various names such as ekub in Ethiopia or tandas in Mexico, involve members contributing to a central pot that is distributed to one member each period until every participant has received the pot once. Despite their global ubiquity, traditional economic analysis often characterizes ROSCAs as suboptimal compared to formal credit markets. This paper addresses the gap between theoretical suboptimality and empirical success by shifting the focus to algorithmic approximation, asking: in the worst-case scenario, how much social welfare is lost due to the informal and decentralized nature of these groups?

Methodology and Framework

The authors model ROSCAs through the lens of algorithmic mechanism design, focusing on two primary variants: Swap ROSCAs, where random turn assignments can be traded in an aftermarket if it benefits both parties, and Auction ROSCAs, where members bid for the pot each round. The analysis utilizes the Price of Anarchy (PoA) framework to measure the efficiency loss of decentralized, self-interested behavior compared to a socially optimal allocation.

To handle the unique structure of ROSCAs—specifically the redistribution of bids as rebates to members rather than disappearing to an auctioneer—the researchers adapted the "smoothness framework" from algorithmic game theory. Beyond theoretical bounds, the study employed computational simulations with synthetic value profiles (Uniform, Pareto, and Unimodal distributions) for groups of 9 and 30 participants. These simulations incorporated Constant Relative Risk Aversion (CRRA) utilities to model how the system performs when participants are risk-averse rather than perfectly rational agents.

Key Results and Robustness

The theoretical analysis yields several constant-factor PoA bounds, demonstrating that ROSCAs are remarkably efficient. Swap ROSCAs are guaranteed a PoA of at most 2.0, ensuring at least 50% of maximum welfare regardless of initial conditions. Sequential second-price and first-price auctions yield bounds of 3.0 and approximately 5.8, respectively.

A critical finding is the "algorithmic robustness" of these systems. The redistribution of bids as rebates acts as a compensatory mechanism, maintaining high welfare even when allocations are not perfectly optimal. Simulation results demonstrate that ROSCAs typically perform much better than their worst-case bounds. As shown in the experimental data, the approximation ratios frequently fall between 1.1 and 1.2, indicating that the ROSCA achieves 80-90% of the theoretical maximum welfare across various parameter settings.

Simulation results showing Price of Anarchy ratios across different parameter settings, consistently falling between 1.0 and 1.2.

Discussion and Future Directions

While the current model provides a strong foundation, it assumes full information and static needs. Real-world ROSCAs often operate under incomplete information and face stochastic shocks, such as sudden medical emergencies, which might change a participant's valuation mid-cycle. Furthermore, social norms, peer pressure, and altruism—which likely enhance the stability and efficiency of these groups—remain areas for future algorithmic exploration. By providing cardinal welfare guarantees, this work offers a new theoretical justification for the enduring success and global prevalence of informal savings circles.

Summary updated

Abstract

Rotating savings and credit associations (roscas) are informal financial organizations common in settings where communities have reduced access to formal financial institutions. In a rosca, a fixed group of participants regularly contribute sums of money to a pot. This pot is then allocated periodically using lottery, aftermarket, or auction mechanisms. Roscas are empirically well-studied in economics. They are, however, challenging to study theoretically due to their dynamic nature. Typical economic analyses of roscas stop at coarse ordinal welfare comparisons to other credit allocation mechanisms, leaving much of roscas' ubiquity unexplained. In this work, we take an algorithmic perspective on the study of roscas. Building on techniques from the price of anarchy literature, we present worst-case welfare approximation guarantees. We further experimentally compare the welfare of outcomes as key features of the environment vary. These cardinal welfare analyses further rationalize the prevalence of roscas. We conclude by discussing several other promising avenues.
BibTeX
@misc{abebe2022algorithmic,
  author = {Rediet Abebe and Adam Eck and Christian Ikeokwu and Samuel Taggart},
  title = {{An Algorithmic Introduction to Savings Circles}},
  year = {2022},
  eprint = {2203.12486},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2203.12486},
}
Figure