English as a foreign language education in East Asia has received repeated criticism for its lack of success in developing sufficient English oral proficiency among its students (Muller at al., 2014). In response to the criticism, the governments of China, Japan and South Korea attempted to include assessment of students’ speaking abilities as part of their high-stakes college entrance exams, hoping for positive washback effects in both primary- and secondary-school English education as well as on shadow education (i.e., non-formal private-sector education). These attempts often failed. In South Korea, a new test called the National English Ability Test (NEAT), which included direct assessment of students’ speaking skill among other skills, was developed in 2012. However, the government's plan to use NEAT to replace the current exam – the Korean College Scholastic Aptitude Test (KCSAT) – was quickly dropped before its implementation. In China, the government has tried to promote more communicative methods of English education through incorporating English speaking test in high-stakes tests such as the Gaokao – college admission tests – in addition to reducing the weight of English in the traditional paper-based exams. However, the policies have received heavy resistance at the regional level and have not been implemented at the national level. In Japan, the government asked universities to accept designated external proficiency tests as part of the Common Test, the existing college entrance exam, in order to make up for the exam's missing speaking component. After a mountain of criticism from test users, implementation of the plan is still pending. In this light, the aim of this paper is to discuss why these policy attempts failed. While these policy attempts occurred in three different contexts, we could see striking underlying commonalities. We argue that these policy attempts were made based on a set of beliefs separate from the reality of the stakeholders (e.g., students, parents and teachers). More specifically, the failures can be largely attributed to the governments’ monolithic view of the English language and their insufficient consideration for equity rather than equality.