In recent years, there has been a growing interest in corporate control issues. Nonetheless, laws governing mergers and inter-firm tender offers were not promulgated until January 2002 in Taiwan. The proxy contest still remains the predominant mechanism for obtaining corporate control and, hence, constructs the focus of the present study. (For an earlier study on proxy contests in Taiwan, see Wei-chao Huang and Gill Yen [1996], Managerial and Decision Economics). Four major changes in proxy rules affect firms listed on the Taiwan Security Exchange from 1984 (the year of promulgation) to 2002 (the year of the latest amendment). Accordingly, the following alternative hypothesis are established: with the exception of the third amendment made on December 17, 1996, the major points of the other three amendments aimed at raising the threshold of minimal shareholdings, lengthening the minimal shaxeholding period, reducing the number of shares to be represented, or making it more costly to send out the proxy materials. Therefore, it is expected that the occurrence of proxy contests will become smaller in the post-amendment period which begins one day immediately after the amendment in question and ends one day earlier than the next amendment. As for the amendment made on December 17, 1996, the forbidding of cash payments cuts both ways. With no pecuniary value, the incentive of shareholders to sign proxy statements is hampered; yet, on the other hand, the costs of soliciting proxies are lowered. Therefore, unlike the other three other amendments, a two-tailed test instead of a one-tailed test is applied to ascertain whether there is a systematic difference in occurrence between pre-amendment and postamendment periods. Corresponding to the first, the second and the fourth rule amendments, as expected, the occurrence ratcas measured by the number of listed companies with proxy contests divided by the number of listed companies--lowered respectively from 0.593 to 0.338, from 0.338 to 0.120, and from 0.202 to 0.140. The differences were statistically significant at least at a 0.10 significance level. For the third amendment, it was found that forbidding cash payments in a company with an increase in the threshold of shareholding, resulted in an increase in the occurence rate from 0.120 to 0.202, which was statistically significant at the 0.05 significance level. (JEL G34)