Wage and compensation vary across individuals, companies, and positions because of different compensation systems. In evaluating the effciency the increased pay from a promotion that is part of these various compensation systems, Rank-Order Tournament Theory states that implementing a compensation scheme based on an individual's relative position within the firm can increase employees' effort and the firm's performance. According to this theory, this compensation structure should have widening pay gaps along the corporate hierarchy. Using data from the Specialty Store Wage and Benefit Survey provided by the National Retail Federation (NRF), this paper tests the existence and the effects of this compensation structure in the retail industry. The dataset contains aggregated information on the number of employees and their average annual salary in each hierarchical level of large national retail chains for the years 1996-1998. With a new approach centering around the second differences between the pay levels, using different statistical tests, including polynomial regression and t-test, we found empirical evidence supporting the existence of a convex compensation structure in the retail industry, where pay level increases at an increasing rate, consistent with the tournament theory. However, through the two-sample t-test and error-in-variables regression, we found that neither the presence of a convex compensation system nor the convexity of this compensation scheme has a statistically significant influence on the firm's performance. The results of this study will solidify the implications of the tournament theory under organizational settings.