This paper examines the conditional risk-return relationship caused by the impact of using realized returns as a proxy for expected returns, which requires a separation of negative and positive market premiums. Following the methodology of Pettengill et al. (1995), we test the cross sectional relationship between beta and realized returns on the Amman Stock Exchange (ASE) for ten beta sorted portfolio over the period of January 1993 to December 2016. The empirical results suggest that the traditional two-pass approach produces an insignificant relationship between beta and realized returns in most of the sample period. However, when adjusting for negative market premiums, the results show a significant and consistent relationship for all the testing periods and samples. However, a guaranteed reward for holding extra risk occurred only in the period 2001 –2008, which suggests an assurance of positive risk-return tradeoff during bull markets. JEL Classifications: G11, G12, G15, C21 Asset Pricing, Emerging Markets, Conditional Relationship, Beta, Market Premium