The English vs. The American Rule on Attorneys Fees: An Empirical Study of Attorney Fee Clauses in Publicly-Held Companies’ Contracts

Abstract

We study attorney fee clauses in a data set of 2,350 contracts contained as exhibits in Form 8-K filings by reporting corporations. Because 8-K filings are required only for material events, these contracts likely are negotiated by sophisticated parties and, therefore, provide evidence of efficient ex ante solutions to contracting problems. The American Rule for compensating attorneys requires each party to pay its own attorney, win or lose; the English Rule (applicable rule in most of the world) requires the losing party to pay the winner’s reasonable fees. Adoption of the English Rule or other loser-pays arrangements has frequently been proposed as a solution to perceived U.S. litigation problems. But the vast theoretical modeling literature on fees has reached no consensus. Empirical reality should help assess the models and provide new insights. Because contracting parties can opt out of the American Rule and into a loser-pays rule at low cost, we expect such opt-outs to occur frequently if the English Rule more efficiently compensates counsel. Our data show, however, that the American Rule is preferred about as often as the English Rule (or similar loser-pays rules). Choosing the American Rule is associated with the following contractual features: specific kinds of contracts, the presence of a non-U.S. party, the absence of arbitration clauses and jury trial waivers, selection of New York law in contracts other than underwriting contracts, and a likely long-term relation between the parties. It is inversely associated with an increasing degree of contract standardization. Sophisticated parties thus often perceive the American Rule to be value-enhancing compared to loser-pays systems but contracting parties that opt out of U.S. courts through arbitration clauses, or eliminate jury trials through jury waiver clauses tend to reject the American Rule. The findings suggest that theoretical models should resist the assumption that a single attorney fee rule is most efficient in all contexts and that models should strive to account for real-world factors associated with fee clauses.