¡@¡@This paper uses the unemployment rate in quarterly data from 1987 to 2011, empirical exploration whether unemployment rate of with Taiwan¡¦s frequent trading countries has a hysteresis effect, as government policy on the basis. ¡@¡@In general circumstances, there are different results by using traditional unit root tests (ADF, PP, KPSS, the DF-of GLS and NP). And the possible reasons are getting the lower power of individual series because of the small simples and without considering about the structural break results in the series with problem of type I error. Therefore, in order to solve the problem, Kyung So, Im, Junsoo Lee, and the Margie Tieslau (2010) constructed an unit root test of panel data which needs to consider the structural break of level and trend. Moreover, the correlation between the inter-individual series by using the cross-sectionally augmented (CA) method from Pesaran (2007) also needs to be considered. ¡@¡@As the results of panel data unit root tests, series of unemployment rate do not reject the null hypothesis, that is, with hysteresis effect which breaks natural unemployment rate continuously. In this case, the government must set up policies to intervene to recover the unemployment rate.