International Business & Economics Research Journal – July 2010 Volume 9, Number 7 107 The Effects Of Unions On Wages By Occupation In The Public Sector John D. Bitzan, North Dakota State University, USA Bahman Bahrami, North Dakota State University, USA ABSTRACT This study examines union wage premiums by occupation in the public sector in the U.S. for the 2000-2004 period. In examining union-nonunion wage differences for public sector workers in occupations accounting for 66 percent of all public workers in the 2000-2004 Current Population Survey, we find positive and statistically significant union premiums for 27 out of 41 occupations examined. We also find large differences among occupations, with miscellaneous teachers and instructors receiving a 61 percent premium, secretaries and administrative assistants receiving a 5 percent premium, and 14 occupations receiving no statistically significant premium. In comparing union premiums by occupation between the private and public sectors, we find, in most cases, that private sector premiums are larger than public sector premiums. Finally, an Oaxaca decomposition shows that the majority of the differential between private sector union premiums and public sector union premiums appears to be due to differences in the way unions reward workers in the private and public sectors, not because of differences in the types of workers in the private and public sectors. Keywords: Public Sector Union Wage Premiums INTRODUCTION ntil recently, most research examining the effects of unions on wages in the U.S. has been in the private sector. However, several recent statistics regarding union membership in the U.S. suggest increased importance of examining these effects in the public sector. For example, while the share of private sector wage and salaried workers that are unionized has declined from 16.8 percent in 1983 to 7.2 percent in 2009, the share of public sector workers that are unionized has remained constant, at 36.7 percent in 1983 and 37.4 percent in 2009 (BLS). Furthermore, although the majority of unionized workers have historically been in the private sector (nearly 68 percent in 1983 according to Blanchflower and Bryson, 2004), the majority are now in the public sector (52 percent in 2009 according to the Bureau of Labor Statistics). These numbers suggest that an understanding of union wage effects requires a better understanding of the effects of unions on wages in the public sector. A recent study by Bahrami, Bitzan, and Leitch (2009) provides several important findings regarding the role of unions in wage determination in the public sector in the U.S. First, there are persistent and consistent union wage premiums in the public sector between 1998 and 2004. Second, union wage premiums measured using a wage equation that allows for different union rewards for different personal characteristics results in higher measured premiums than those obtained with a traditional union wage equation that only measures the premium with an intercept shift. Third, there are substantial differences in union premiums for different types of workers – e.g. local workers receive higher union premiums than state workers, who in turn receive higher union premiums than federal workers; blue collar workers receive higher union premiums than white collar workers; part time workers receive higher union premiums than full time workers; female workers receive higher union premiums than male workers; and black workers receive higher union premiums than white workers. Finally, a large difference between the union wage premium in the private and public sectors still exists (22.7 percent in the private sector and 11.2 percent in the public sector), and much of it is explained by differences in public and private sector worker characteristics (40 to 50 percent). Specifically 22 to 30 percent of the difference in private-public sector union premiums is due to a much larger proportion of public sector workers in white collar occupations in comparison to the private sector, and 17 to U