An Empirical Investigation of Settlement and Litigation—The Case of Taiwanese Labor Disputes


  • We thank Yi-Jan Yu, Shih-Wen Haung, Yen-Chi Huang, Pei-Ru Kao, Chen-Chieh Ting, and Tai-Hsiang Wang for invaluable research assistance. We are indebted to Kevin M. Clermont for his thoughtful suggestions. For comments, we thank Theodore Eisenberg, Tze-Shiou Chien, Mostafa Beshkar, and participants at the 2009 Conference on Empirical Legal Studies. We also thank the Center for Institution and Behavior Studies and Research Center for Humanities and Social Sciences as well as Institutum Iurisprudentiae, Academia Sinica, for funding support.

Kuo-Chang Huang, Associate Research Professor, Institutum Iurisprudentiae, Academia Sinica, 128 Academia Sinica Rd., Sec. 2, Nankang, Taipei 11529, Taiwan; email: Chen is Research Professor, RCHSS, Academia Sinica; Lin is Assistant Research Professor, Institute of Economics, Academia Sinica.


Using newly collected data on labor disputes in Taiwan, we study the characteristics of cases settled out of court, cases dropped after settlement fails, and cases actually brought to court. We find that the size of the stakes plays a crucial role in routing between settlement and litigation. In particular, our study shows that increasing stakes decreases the probability of settlement, decreases the percent of the claim recovered in settlement, and increases the probability of litigation after settlement fails. Moreover, while the plaintiff's credibility of threatening a lawsuit is an important factor, its absence is not so important as to undermine the pattern that an increase in the stakes would decrease the likelihood of settlement. Our study also reveals that the mediation mechanism performs the function of allowing workers with small claims to obtain effective recovery, which would be impossible if those workers had to resort to litigation. This result also suggests that the traditional game-theoretic models are inadequate in explaining the litigants' behavior in settlement of small claims. Finally, we find that the workers' risk aversion might cause the percent of the claim recovered in settlement to drop as the amount of their initial claim increases.