Introduction

For much of the postwar period, a higher share of African-American workers have been in unions than workers from other racial and ethnic backgrounds. As union representation and union coverage have declined for the country as a whole, unionization rates for African-Americans have fallen more quickly than for the rest of the workforce. Black workers are still about 30 percent more likely than the rest of the workforce to be in a union today, but as recently as the mid-1980s, black workers were almost 50 percent more likely to be in a union or covered by a union at their workplace. Part of the reason for the decline in unionization rates among African-Americans is undoubtedly related to the decline of U.S. manufacturing. For example, since the 1960s, African-Americans were more likely to work in the heavily unionized automotive sector than white or Latino workers. As these sectors have declined in relative importance, unionization rates for blacks have also dropped. The overall decline in manufacturing, however, is only part of the problem. First, since the early 1990s, the share of black workers in manufacturing has been falling more rapidly than the manufacturing share for the workforce as a whole. From the end of the 1970s through the early 1990s, African-Americans were just as likely as workers from other racial and ethnic groups to have manufacturing jobs. Since the early 1990s, however, black workers have lost considerable ground in manufacturing. By 2006, blacks were about 15 percent less likely than other workers to have a job in manufacturing. Second, even within manufacturing, unionization rates have been on the decline, to the point where manufacturing workers now are no more likely to be in a union than workers in the rest of the economy (see Schmitt and Zipperer, 2007a). Meanwhile, over the last 25 years, unionization rates have held steady in the public sector, which suggests that employer opposition to unions, not simply economic restructuring, lies behind the decline in overall unionization rates (see Schmitt and Zipperer, 2007b).

Findings

Our analysis of data from the Current Population Survey, the government's most important regular source of data on the labor market, also finds:

• Between 1983 and 2006, the share of African-American workers who were either members of a union or represented by a union at their place of employment fell substantially, from 31.7 percent of all black workers in 1983 to 16.0 percent in 2006. In 2006, African-Americans were still more likely to be in a union (16.0 percent) than whites (13.3) and Hispanics (10.7 percent). Nevertheless, the decline in union membership for black workers between 1983 and 2006 was sharper for blacks (down 15.7 percentage points) than it was for whites (down 8.9 percentage points) or Hispanics (down 13.5 percentage points). (See Table 1.)

• The share of African-Americans working in manufacturing declined from 23.9 percent in 1979 to 10.1 percent in 2006. Whites saw slightly smaller declines (from 23.5 percent to 11.9 percent), while Hispanics experienced a bigger drop (from 30.2 percent to 12.6 percent). (See Table 2.)

• Between 1979 and 2006, the share of workers in auto manufacturing dropped for blacks, whites, and Hispanics. Blacks suffered the biggest decline, a 0.8 percentage-point decline, from 2.1 percentage points in 1979 to 1.3 percentage points in 2006, compared to a 0.3 percentage point drop for whites and Hispanics. (See Table 3.)

• Throughout the entire 1983-2006 period, black workers have made up 13-15 percent of all union workers. Over the same period, the share of whites in the total union workforce fell from 78.1 percent to 69.2 percent, while the share of Hispanics rose from 5.8 percent to 11.5 percent of all union workers. (See Table 4.)

• Throughout the entire period from 1979 to 2006, the share of African-American workers in the total manufacturing workforce hovered around 10 percent. Meanwhile, white workers fell from 82.7 percent of all manufacturing workers in 1979 to 69.9 percent in 2006. Hispanics (and other workers) significantly increased their representation in the manufacturing workforce over the same period (up from 6.0 percent of manufacturing jobs in 1979 to 14.5 percent in 2006, for Hispanics). (See Table 5.)

• In 2006, blacks made up 14.1 percent of the auto manufacturing workforce; whites were 73.3 percent; Hispanics, 7.6 percent. Between 1979 and 2006, the share of blacks and whites in the total auto manufacturing workforce both fell. The share of Hispanic and other workers, meanwhile, increased, although remained at relatively low levels. (See Table 6.)

• Interpreting the preceding changes in the distribution of African-American employment and unionization rates can be complicated since the share of African-Americans and other racial and ethnic groups in the total workforce has changed over time. Tables 8, 9, and 10 show the "relative representation" of whites, blacks, and Hispanics in the total union workforce (Table 8), the total manufacturing workforce (Table 9), and the auto manufacturing workforce (Table 10). The simplest way to explain the calculation is with an example. To calculate the relative representation of African-Americans in the share of all union workers, we take the ratio of the share of African-Americans in all union workers (14.0 percent in 2006, in Table 4) to the share of African-Americans in the total workforce (10.8 percent in 2006, in Table 7). The resulting ratio is 1.30, which is greater than 1, indicating that African-Americans are "over-represented" among union workers since there is a larger share of African-Americans in unions than there is in the workforce as a whole. In the same year, Hispanics made up 11.5 percent of union workers, but 13.6 percent of the total workforce, resulting in a relative representation of 0.85, which is less than 1, indicating that Hispanics are "under-represented" among union workers. If a group has the same share of workers in unions as they do in the total workforce, then the ratio for relative representation would equal one (which is close to the rate for whites in 2006, 0.99).

• In 2006, African-American workers were "over-represented" in unions (Table 8) and in auto manufacturing (Table 10), but are actually "under-represented" in manufacturing as a whole. (See Table 9.)

• The relative representation of African-Americans has been slowly declining in unions (Table 8), manufacturing (from the mid-1990s, Table 9), and auto manufacturing (Table 10). From the mid-1990s on, black workers have actually been under-represented in manufacturing, relative to the rest of the economy (Table 9). Meanwhile, white workers have held their ground in manufacturing (Table 9), and have actually increased their relative representation in unions (Table 8) and auto manufacturing. (See Table 10.)

John Schmitt is a senior economist and Ben Zipperer is a research assistant at the Center for Economic and Policy Research in Washington, D.C. Click here for a PDF version of the complete report.

References

Center for Economic and Policy Research. Uniform Extracts of the Current Population Survey Outgoing Rotation Group. 2007.

Schmitt, John and Ben Zipperer. 2007a. "Union Rates Fall in 2006, Severe Drop in Manufacturing," Washington, D.C.: Center for Economic and Policy Research Union Byte (January).

Schmitt, John and Ben Zipperer. 2007b. "Dropping the Ax: Illegal Firings During Union Election Campaigns," Washington, D.C.: Center for Economic and Policy Research Briefing Paper (January).