Abstract
The Maker Protocol is a decentralized finance application that enables collateralized lending. The application uses open-bid, second-price auctions to complete its loan liquidation process. In this paper, we develop a bidding function for these auctions, focusing on the costs incurred to participate in the auctions. We then optimize these costs using parameters from historical auction data, and compare our optimal bidding prices to the historical auction prices. We find that the majority of auctions end at higher prices than our recommended optimal prices, and we propose several theories for these results.
Introduction
In the Decentralized Finance (DeFi) ecosystem, the Maker Protocol allows users to secure loans in the DAI stablecoin by locking up collateral assets. When the value of this collateral falls below a safety threshold, the protocol initiates liquidation auctions. While these auctions appear to be straightforward opportunities for automated "keepers" to purchase assets at a discount, the process involves significant hidden costs. Participating in a blockchain auction requires paying "gas" fees, incurring "slippage" during asset swaps on decentralized exchanges, and tying up capital that could otherwise earn interest. This paper establishes a mathematical framework to define the "true" optimal bid by accounting for these frictions and evaluates whether real-world participants act rationally.
Modeling Participation Costs
The researchers defined a bidding function where the expected bid price is determined by the asset's market value, private utility, and three primary participation costs: transaction fees (gas), conversion costs (slippage), and the opportunity cost of capital.
To ground this model in reality, the authors benchmarked gas usage using a keeper bot on the Kovan testnet and analyzed historical Etherscan data for common transaction types such as joining, exiting, and trading. This empirical approach allowed them to quantify the fixed and variable costs associated with each auction event.
Methodology and Optimization
The study utilized data from 155 Maker auctions completed between March and July 2020, alongside historical Uniswap reserve data. Using MATLAB, the authors formulated a convex optimization problem to minimize total participation costs. They focused on two controllable variables: maximum portfolio size ($V_{max}$) and the rebalance margin ($R$). This optimization allows for the calculation of a "theoretical minimum auction discount"—the point at which a bid becomes rational after all costs are considered.
Results: The Reality of Overbidding
A key finding of the study is the significant "friction" associated with small bids. Because many participation costs, like gas fees, are fixed or semi-fixed, bidders on small lots (under $1,000) must demand much higher discounts to remain profitable compared to those bidding on large lots.
When comparing this theoretical optimum to historical data, the results were startling: 75% of analyzed auctions ended at prices higher than the calculated optimal price. This overbidding was even more pronounced in small lots, where 83% of bids exceeded the rational threshold. Even when excluding the cost of capital, 74% of participants appeared to "overpay" relative to the theoretical optimum.
Discussion and The Altruism Hypothesis
The prevalence of overbidding suggests that Maker auctions are "interdependent-value" auctions, combining common-value components (market price) with private-value constraints (individual gas and capital costs). The authors propose the Altruism Hypothesis: many participants may be "Altruistic Actors"—individuals or organizations with a vested interest in the protocol's survival. By bidding at a loss, these actors prevent "zero-bid" disasters and ensure system solvency, particularly during periods of extreme market volatility.
Conclusion
While the current model assumes a static win probability and ignores certain off-chain overheads, it highlights a significant gap between theoretical rationality and observed on-chain behavior. As the Maker Protocol transitions to a Dutch auction system, the dynamics of participation costs will shift, potentially incorporating new factors like flash loans. Understanding these costs and the behavioral drivers behind participant altruism remains essential for maintaining decentralized financial stability.
Summary updated