XII. LABOR STUDIES/LABOR UNIONS,
COLLECTIVE BARGAINING, DISPUTE
RESOLUTION, AND LABOR AND
EMPLOYMENT LAW REFEREED PAPERS

Union Organizing Commitment:
Rhetoric and Reality¹

Jack Fiorito
Florida State University

Paul Jarley
University of Kentucky

Abstract

     This paper examines union organizing effectiveness and effort among national unions in the United States. Indicators are compared for 1986-90 and for 1996-98. Despite considerably more rhetoric favoring organizing since the 1995 election of John Sweeney to head the AFL-CIO, the data indicate modest or small changes in the average national union's organizing effort or success. Although some "shuffling of the ranks" has occurred, generally, unions that emphasized organizing in the earlier period also tended to emphasize organizing in the latter period. Considerable variation among national union efforts and success is noted, and future research on the reasons for these differences is encouraged.

     When John Sweeney challenged AFL-CIO leadership in 1995, many union leaders felt the incumbent Kirkland-Donahue leadership was offering too little too late. Oversimplified, Sweeney's challenge was "Organize!" Sweeney's campaign included other elements, but the centerpiece was a pledge to boost organizing, challenging unions to spend 30 percent of their budgets on organizing by 2000. It is unclear how much unions were then spending, but estimates of 3 percent have been offered. Just how this dramatic shift was to be accomplished was not entirely clear. It is also not clear just how far unions have since moved in the direction of the new organizing emphasis, although rhetoric in support of organizing has clearly increased. This paper examines evidence on organizing efforts and results from before and after 1995, and provides an assessment of union efforts to refocus on organizing nonunion workers.

National Union Organizing Efforts and Results

     Organizing-related data for large unions are presented in Table 1. These 25 unions are the largest of approximately 100 national unions. In the first two columns, union size is shown in thousands of members. Collectively, these unions claim about 13.7 million members, or roughly 85 percent of U.S. membership. The next three columns provide percentage growth rates. First, growth is shown for 1990-97 and is based on the size figures shown. Growth is also shown for 1989-91 and 1995-97.

     Membership change reflects many factors, including mergers, retirements, quits, closures, downsizing, and organizing. Though growth is a "bottom line," it is by no means a clear indicator of organizing activity or success. It is but one piece of evidence (Fiorito et al. 1995). That said, this evidence indicates substantial variation across unions and time. Unions are about evenly divided between growth and shrinkage in any of the three periods. The average growth rate for the unions shown was .4 percent in 1989-91 and -.5 percent in 1995-97 (.2 percent each period for all unions).² Only about half show consistent growth or decline over all three periods, and among these, variation in rates is sometimes dramatic--the SEIU grew more than 10 times faster in 1989-91 than in 1995-97. Still, the correlation between growth rates for these two periods is .48 (p < .05), indicating some temporal consistency (but r = .03, NS [not significant], for all unions). Again, caution is advised in attaching too much meaning to any particular figure. For example, the rapid SEIU or IBT growth of 1989-91 may indicate new affiliations or change in per capita payment basis rather than organizing.

     The next four columns provide win rates (in percent) in NLRB elections and union officials' self-ratings (four-point scale) of their union's organizing effectiveness in the late 1980s and mid-to-late 1990s. Note that organizing by some unions (e.g., NEA and AFT) takes place mainly outside NLRB jurisdiction. In addition, several unions make efforts to avoid NLRB election processes because they see them as prone to employer manipulation. Still, NLRB win rates provide partial indication of organizing effectiveness and show some consistency over time. The correlation for the NLRB win rates shown is .46 (p < .10; .43, p < .05, for all unions). There appear to be persistent union-specific differences. For example, the IBT has relatively low win rates, and the SEIU has relatively high rates in both periods. These may indicate different organizing strategies--the IBT casting a broad net and accepting low success rates, the SEIU focusing efforts more narrowly with greater success (see Fiorito et al. 1995). Six of the 18 unions shown with win rates for both years show changes of 10 percentage points or greater, with five of these showing increased success. This seems to indicate improved efficiency in winning, or more focused efforts. For the 18 unions together, the success rate improved from 51 percent to 53 percent (from 49 percent to 52 percent for all unions).

     There is less temporal consistency for the unions shown in organizing effectiveness self-rating (r = .09, NS; but r = .41, p < .01 for all unions). For the unions shown, the average rating dropped from 3.3 to 3.1. For all unions, the drop was from 2.9 to 2.5. The correlation between NLRB win rates and self-rated organizing effectiveness for unions shown is 0.75 (p < .01) for 1990, but only 0.18 (NS) for 1997 (r = .50, p<.01, for 1990; and r = .28, NS, in 1997 for all unions). Thus, NLRB win rates seem to enter leader assessments of organizing effectiveness, although less clearly so recently, possibly reflecting greater emphasis on "NLRB avoidance" or organizing effort in effectiveness assessments. The membership growth rate for 1989-91 correlates insignificantly with the 1986-90 NLRB win rate (r = .26, NS) and the 1990 self-rating (r = .33, NS), whereas the 1995-97 membership growth rate correlates significantly with the 1996-98 NLRB win rate (r = .47; p < .10), but not with the 1997 self-rating (r = .', NS). (For all unions, the respective relations for the earlier period are r = .12, NS, and r = .04, NS; and for the latter, r = .43, p < .05, and r = .21, NS.) All told, the three indicators (membership growth, NLRB win rate, and effectiveness self-rating) show some, but not always strong, consistency over time and across measures. Each indicates organizing effectiveness, but none is clearly sufficient and reliable. In sum, they show membership growth is low and little changed in the last 15 years, NLRB win rates are up slightly, and self-assessed organizing effectiveness is down slightly.

     The next four columns of figures relate more clearly to organizing effort than to effectiveness. The two columns labeled "Organizing Effort" refer to the number of workers the union has sought NLRB recognition for divided by union membership (in thousands)³ and thus represents organizing effort relative to resources (proxied by membership). Twelve of the 25 unions show increases in organizing effort, nine show decreases, and four lack comparable data. For the unions shown, the average in the earlier period was 58.4 versus 61.9 for the later period. (For all unions, the respective averages were 66.4 and 63.2, but the comparison is less meaningful because of sample composition change.) It thus appears that a slight increase in relative organizing effort has occurred. Despite some changes, unions tended to maintain their respective effort levels in both periods as indicated by a .91 (p < .01) correlation between the organizing effort figures for the two periods (r = .68, p < .01 for all unions).

     In 1997, unions were questioned on their commitment to organizing (relative to other unions) and budget percentages devoted to organizing (Fiorito et al. 1998). Large unions see themselves as more committed to organizing. Their average score is 3.2, versus 2.8 for all unions. (Two of the four large unions "disagreeing" enjoy very high union density, and organizing may not be a high priority for them.) This expressed commitment is mildly borne out in comparison to the relative organizing effort measure. Although the correlation between relative effort and stated organizing commitment is insignificant among the large unions shown (r = .14), for all unions the correlation is positive (r = .32, p < .10).

     A problem with the organizing budget data is that many unions did not--perhaps could not--provide estimates. Nonetheless, on average, 21.3 percent of budgets was reportedly spent on organizing for the large unions shown, but for all unions the figure was 15%.⁴ The higher figure for large unions is not echoed in the relative organizing effort data (r = -.50; r = .13 for all unions; both NS, with samples of only 8 and 20, respectively). The budget figures, however, do align with respondents' expressed organizing commitment (r = .64, p < .10; r = .58, p < .01 for all unions).

     Fiorito et al. (1995) noted that indicators of organizing effectiveness and effort tended to form two clusters. For the NLRB-based effort measure, that observation tends to hold for both periods examined here, with correlations suggesting a negative or no relation to all three organizing effectiveness indicators examined earlier. A slightly different picture emerges when comparing effectiveness indicators with the organizing commitment and budget share measures. Relations range from "none" to positive, so assessments of the relation may be sensitive to the effort measure considered.

     The final two columns in Table 1 look at "representational specialization"--the percent of organizing each union undertakes within its traditional industry jurisdiction. Superficially, it may seem many unions have shifted their organizing strategies to "targets of opportunity" based on frequent reports of, for example, miners organizing nurses and auto workers organizing writers. Yet the data show no real trend. Representational specialization increased and decreased for roughly equal numbers of unions.

     It would be foolhardy to assert that this review provides a complete picture. Some of the most critical issues are neglected or barely touched, such as the extent to which unions have adopted the "organizing model" (OM) and information technology (IT) as organizing tools. As Heery et al. (2000) have shown for the United Kingdom, one can measure union adoption of OM elements, but ironically there are no systematic U.S. data on OM adoption. The data on organizing commitment are also limited. There has been no published U.S. study systematically linking either OM adoption or organizing commitment to organizing outcomes, except Bronfenbrenner (1997), who examined the impact of organizing tactics that can be linked with OM concepts. There is as yet no published study linking union IT use to organizing outcomes. Despite limitations, it is clear that union organizing efforts and results vary substantially across unions, even in leaders' assessments, where social desirability biases could skew responses.

Discussion

     Various writers have suggested a need to fundamentally rethink union goals, strategies, and structures, in some instances suggesting new models or resurrecting old ones (see Turner et al. 2001). There appears to be consensus in some quarters that improving efficiency and enhancing organizing commitment of existing unions will not suffice (although prescriptions then diverge). Heckscher notes: "The focus on organizing by the AFL-CIO in the last few years has clearly not uncapped a powerful wellspring of desire for unionization" (2001:59). Data reviewed here suggest that, although the focus is more squarely on organizing, the resource reallocations have not yet matched the rhetoric. It may be premature to conclude that current unionism models cannot meet the challenge. It is not clear they have been implemented, that commitments and efforts to organize within those models have been forthcoming.

     The evidence reviewed here and more formal modeling efforts underscore that union strategies, structures, and tactics matter. As yet, our ability to draw strong inferences and policy implications are quite limited. Efforts to model phenomena such as union organizing activity levels have produced limited explanatory power (e.g., Voos 1987). Even so, it appears clear that U.S. unions that are innovative, advanced in IT use, decentralized, committed to organizing, strategic in selecting targets, and using OM tactics are faring better than others in organizing (Bronfenbrenner 1997; Fiorito et al. 1995, 2002).

     Yet it is hard not to feel that something more is needed. Even unions that have led the "transforming to organize" movement have failed to score consistently impressive gains. The AFL-CIO has set a goal of organizing one million new members annually (Lazarovici 2001). At expected membership attrition and workforce expansion, this would restore meaningful growth in union density. In 2001, AFL-CIO unions reportedly added almost 500,000 members (AFL-CIO 2002), but the net gain in U.S. membership was just 17,000, and union density held steady at 13.5 percent (U.S. Department of Labor 2002). To be sure, some of these numbers represent improvement over recent years and are all the more impressive given losses of nearly 200,000 union manufacturing jobs in 2001.

     But they also gloss over some conspicuous failures, including large organizing campaigns among Delta flight attendants and Nissan autoworkers that resulted in decisive union losses. In both cases, unions may justly cite unfair and/or illegal employer campaign tactics, but this is little solace in view of unlikely reform prospects (e.g., tougher laws or voluntary employer behavior change). The Nissan case is particularly troubling, because it was the fourth UAW attempt to organize these workers, and the UAW seems to have made little progress, gaining only one-third of the votes, the same support won in its 1989 organizing drive (Hakim 2001). Some might attribute these failures to insufficient OM adoption. At least in the Nissan case, there is suggestion that the UAW tried a new approach placing less emphasis on developing inside support and worker-to-worker recruiting, although some parts of its approach were clearly "OM-compatible."

     Bronfenbrenner (2001) asserts that unions already know what they must do but that getting diverse autonomous national unions to change is the real challenge. Although many are less certain about what is needed to organize and more impressed by the organizing challenge, there is consensus about the difficulty of effecting organizational change (e.g., Fletcher and Hurd 2001). Calls for decentralization and innovation do not always find receptive audiences among central authorities (Craft 1991; Delaney et al. 1996). Organizational change can be difficult even in top-down businesses. The intertwined administrative and representative systems of unions increase the complexity. There is a difficult balance to achieve. Although decentralization has appeal in many regards, the support and expertise of a competent national headquarters operation can be vital (Fiorito et al. 1995; Voss and Sherman 2000). This is not to say that there is a "magic formula" for all unions. U.S. unions face diverse challenges in their environments (Katz 2001). Not all members or prospective members find the OM appealing. White-collar or professional workers may prefer a more service-oriented model. Cuts in servicing to pay for organizing may anger members who perceive they are getting less value for their union dues. Workers generally, and perhaps U.S. workers particularly, already feel intense time pressure. Who has time to undertake unpaid work for the union with so many other demands already at hand?

     Others have suggested that the OM is not a proper "model," in that it fails to specify a clear path to organizing nonunion workers as the objective. In this vein, Hurd (1998) has proposed making the transformation to organizing unions (or cultures) the prime objective (also see Fletcher and Hurd 2001). As Fiorito et al. (1991) note, however, although organizing deserves special prominence as a strategy to serve virtually all union objectives, it is not a goal itself. More attention is needed to concepts such as Masters and Atkin's (1999) "value-added unionism," which recognizes that there are alternative paths by which unions can add value for workers, employers, and society. In some contexts, adding value may favor direct benefits and services to members or other forms of mutual aid, in others "partnership" with employers and in still others an emphasis on political action, "community unionism," or workplace activism. Heery et al.'s results (2002) suggest that British unions' organizing outcomes are linked positively with various strategies.

     National union autonomy emerged and persists as a central organizational imperative for U.S. labor because workers face highly diverse work environments with different needs and challenges not easily well served via "one big unionism." There are indeed common "transunion" interests that shift ground over time, and interunion collaborations and central federations address them. But too much emphasis on these at the expense of national union autonomy evokes the folly and frustration of herding cats. Still, there is much central federations can do to assist affiliates.

     We can surely learn more about what works in diverse union environments and about what works in particular environments through cross-union studies. Simultaneously, we should recognize limitations of "thinking inside the box." Focusing on existing institutions and their capabilities to represent workers is limiting. In economists' terms, perhaps there is too much emphasis on the supply side. Heckscher's (2001) comment about the apparent failure of the organizing refocus to unleash a "wellspring" of latent unionism is apt. Although there is some indication that refocusing has stimulated further change, new representation forms or new union structures may better tap this potential wellspring. Here, too, central bodies can play a critical role, providing focal points for community unionism or vehicles for experiments. Recall that it took a "radical" CIO to help unleash a torrent of industrial unionism in the face of AFL craft unions' resistance in the 1930s. The CIO was not only about structure; it also entailed a more "social unionism." This torrent was aided by an attitudinal and legal environment uniquely reshaped by the Great Depression. One can see similar, smaller-scale attitude-shifting potential in events such as the Enron scandal or even the September 11 terrorist attacks, but whether and how such shifts occur is unpredictable. It is also difficult to foresee whether current unions (or new worker representation forms) can provide a vision that will inspire workers and turn latent union support into dramatic organizing gains.

Notes

     ^1. This paper draws heavily from a chapter by the first author in the forthcoming volume edited by Gregor Gall, Union Organizing, London: Routledge, 2003. This material is used here with the permission of the editor and publisher. The authors thank Gregor Gall for editorial suggestions on that earlier work. In addition, Court Gifford of BNA and various staff members at the U.S. Department of Labor and the National Labor Relations Board provided assistance in obtaining data. Union staff and officers who responded to the Survey of Union Information Technology are also due a sincere note of gratitude.

     ^2. Throughout this passage, results are reported for the large unions shown in Table 1, and separately for all unions for which data are available. Data availability and sample composition varies with time periods due to mergers, nonresponse, etc.

     ^3. The 1986-90 period consists of 54 months ending in June 1990, while the latter consists of the 36 months of the calendar years 1996-98. For comparability, the numerator for the latter period is multiplied by 54/36.

     ^4. The figure for organizing budget percentage in Table 1 is actually an average of three responses corresponding to 1996, 1997 (estimated), and 1998 (estimated) figures for the union. The respective averages are 12 percent, 14 percent, and 16 percent. Thus one could say that unions were moving in the direction called for by Sweeney, but the trend line would obviously not have reached 30% in 2000 based on these figures. The three-year average figure of 15% given is only for unions supplying estimates for all three years.

References

AFL-CIO. 2002. Work in Progress (January 6). Online at www.aflcio.org/aboutaflcio/wip/wip01062003.cfm.

Bronfenbrenner, K. 2001. "Changing to Organize," Nation (September 3). Online at www.thenation.com/docPrint.mhtml?i=20010903ands=bronfrenbrenner.

Bronfenbrenner, K. 1997. "The Role of Union Strategies in NLRB Certification Elections." Industrial and Labor Relations Review, Vol. 50, pp. 195-212.

Craft, J.A. 1991. "Unions, Bureaucracy, and Change: Old Dogs Learn New Tricks Very Slowly." Journal of Labor Research, Vol. 12, pp. 393-405.

Delaney, J.T., P. Jarley, and J. Fiorito. 1996. "Planning for Change: Determinants of Innovation in U.S. National Unions." Industrial and Labor Relations Review, Vol. 49, pp. 597-614.

Delaney, J.T., J. Fiorito, and P. Jarley. 1991. Union Innovation and Effectiveness: Results from the National Union Survey. Iowa City: University of Iowa College of Business, Industrial Relations Research Institute.

Fiorito, J., W. E. Hendricks, and C. L. Gramm. 1991. "Union Structural Choices," in G. Strauss, D. G. Gallagher, and J. Fiorito, eds., The State of the Unions. Madison, WI: Industrial Relations Research Association, pp. 103-37.

Fiorito, J., P. Jarley, and J. T. Delaney. 2002. Information Technology, Union Organizing, and Union Effectiveness. Working paper, Florida State University.

Fiorito, J., P. Jarley, and J. T. Delaney. 1998. Luddites, IT, and Cyberunions: Information Technology and National Unions. Report on the Survey of Union Information Technology, Florida State University.

Fiorito, J., P. Jarley, and J. T. Delaney. 2001. "National Unions as Organisations." In G. R. Ferris, ed., Research in Personnel and Human Resources Management, Vol. 20, pp. 231-68.

Fiorito, J., P. Jarley, and J. T. Delaney. 1995. "National Union Effectiveness in Organizing: Measures and Influences." Industrial and Labor Relations Review, Vol. 48, pp. 613-35.

Fletcher, B., Jr., and R. W. Hurd. 2001. "Overcoming Obstacles to Transformation: Challenges on the Way to a New Unionism." In L. Turner, H. C. Katz, and R.W. Hurd, eds., Rekindling the Movement. Ithaca, NY: ILR Press, pp. 182-208.

Gifford, C. D., ed. 2001. Directory of US Labor Organizations, 2001 edition. Washington, DC: BNA Books.

Hakim, D. 2001. "Big Loss at Nissan Seems to Undercut UAW Objectives," New York Times on the Web. Online at www.nytimes.com/2001/10/05.

Heckscher, C. C. 2001. "Living with Flexibility." In L. Turner, H. C. Katz, and R. W. Hurd, eds., Rekindling the Movement. Ithaca, NY: ILR Press, pp. 59-81.

Heery, E., M. Simms, R. Delbridge, J. Salmon, and D. Simpson. 2002. "Trade Union Recruitment Policy in Britain: Form and Effects." In G. Gall, ed., Union Organizing and Trade Union Recognition in Britain. London: Routledge, in press.

Heery, E., M. Simms, D. Simpson, R. Delbridge, and J. Salmon. 2000. "Organizing Unionism Comes to the UK." Employee Relations, Vol. 22, pp. 38-57.

Hurd, R. W. 1998. "Contesting the Dinosaur Image: The Labor Movement's Search for a Future." Labor Studies Journal, Vol. 22, pp. 5-30.

Katz, H. C. 2001. "Afterword: Whither the Labor Movement?" In L. Turner, H. C. Katz, and R. W. Hurd, eds., Rekindling the Movement. Ithaca, NY: ILR Press, pp. 339-49.

Lazarovici, L. 2001. "Bigger, Faster, Better Organizing." America@Work, Vol. 6, no. 1 (January), pp. 8-11.

Masters, M. F., and R. S. Atkin. 1999. "Union Strategies for Revival: A Conceptual Framework and Literature Review." In G. R. Ferris, ed., Research in Personnel and Human Resources Management, Vol. 17, pp. 283-314.

Turner, L., H. C. Katz, and R. W. Hurd, eds. Rekindling the Movement. Ithaca, NY: ILR Press.

U.S. Department of Labor, Bureau of Labor Statistics. 2002. Union Members in 2001. Online at http://www.bls.gov/cps/.

Voos, P. B. 1987. "Union Organizing Expenditures: Determinants and Their Implications for Union Growth." Journal of Labor Research, Vol. 8, pp. 19-30.

Voss, K., and R. Sherman. 2000. "Breaking the Iron Law of Oligarchy: Union Revitalization in the American Labor Movement." American Journal of Sociology, Vol.106, pp. 303-49.