Loopholes in Complex Contracts

Kenneth Ayotte & Adam B. Badawi

American Law and Economics Review2025https://doi.org/10.1093/aler/ahaf005article
AJG 2ABDC B
Weight
0.37

Abstract

In 2016, J. Crew exploited a loophole in a complex debt contract, allowing it to remove $250 million of collateral from its lenders’ reach. How could a flaw like this arise in a contract between sophisticated commercial actors? Answering this question requires a model where contracts can be imperfect and evolving. We create such a model in a principal-agent setting using genetic algorithms, a tool from the complexity science literature. As in real-world practice, contracts in a genetic algorithm evolve from prior contracts, combining terms from the “parent” generation based on their past performance. Our main result is that loopholes arise as a byproduct of evolutionary learning. Through experience, contracting parties gradually allow the agent’s activity level to increase as they learn how to permit more activity without increasing agency costs. But, as activity rises, contracts become more complex in two ways. First, as the effective length of the contract increases, more value-relevant states of the world arise over time. This increases the number of potential mistaken terms that the agent can exploit. Second, the interactivity of terms increases, making loopholes more severe. As contracts evolve, then, they can achieve greater performance but also become less robust to mistakes. (JEL K12, G34)

1 citation

Open via your library →

Cite this paper

https://doi.org/https://doi.org/10.1093/aler/ahaf005

Or copy a formatted citation

@article{kenneth2025,
  title        = {{Loopholes in Complex Contracts}},
  author       = {Kenneth Ayotte & Adam B. Badawi},
  journal      = {American Law and Economics Review},
  year         = {2025},
  doi          = {https://doi.org/https://doi.org/10.1093/aler/ahaf005},
}

Paste directly into BibTeX, Zotero, or your reference manager.

Flag this paper

Loopholes in Complex Contracts

Flags are reviewed by the Arbiter methodology team within 5 business days.


Evidence weight

0.37

Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40

F · citation impact0.16 × 0.4 = 0.06
M · momentum0.53 × 0.15 = 0.08
V · venue signal0.50 × 0.05 = 0.03
R · text relevance †0.50 × 0.4 = 0.20

† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.