Higher union density (the percentage of employees in an area who belong to unions) is known to raise the wages of union members. We find that in the supermarket industry, higher density locally also leads to higher wages for non-union members. Despite this, workers who are not in unions lose ground relative to union members. For a 10% increase in local union densities in the supermarket industry it is estimated that the wages of union employees in that labor market will increase by 5.3% and by 1.2% for nonunion. Hence, the union wage premium will increase in regions with higher union densities. At the time of the last national examination of the retail industry in 1977, union influence in the retail food industry was near its peak. Subsequently union membership and influence have declined. In 1993 a reported 25.7% of supermarket employees listed themselves as union members, a decline from 29.4% in 1984 (NBER, 1995). After conducting an extensive review of wage estimation literature, we focus on the structure of wages in the supermarket industry between 1984 and 1993. The effect of union penetration in local markets on nonunion wages or the "spill-over" effect, is an important focus of this study. While it is widely accepted that the higher the union percentage in local labor markets the higher the wages for those in the union, the corresponding effect has not been closely examined for the nonunion sector. To investigate this phenomenon we first replicate as closely as possible the results of a previous supermarket wage study (Belman and Voos, 1993). Following a baseline comparison of this initial estimation, we enlarge the data set to include individuals in rural areas and from additional years. By using a more sophisticated estimation technique we find that increasing union densities positively affect union and nonunion wages. The significant results for the nonunion sector is of particular importance, as this is the first confirmation of this effect to the best of our knowledge.