One Voice,But Whose Voice? Exploring What Drives Trade Association Activity

Abstract

Trade associations operate under the premise of advancing the shared interests of their member firms. How well do they fulfill this role? This article measures the activity of 148 major industry trade associations over time and relates this activity to the performance of the relevant industries and dominant firms within them. Findings suggest that trade association spending increases when the profitability of the four largest firms in an industry decreases, but spending is unrelated to the profitability of the industry overall. This implies that large firms exert control over trade association agendas and may use these communal organizations to advance their own interests rather than the shared interests of the entire industry. Moreover, it points to the need for further development of the currently anemic management literature on the activities of trade associations.

Keywords

trade associations collective action cooperative strategy industry effects

Trade associations are a pervasive part of our economic landscape. Most firms hold membership in at least one trade association. Nearly every industry, across all sectors of the economy, has organized at least one and often several trade associations. From the American Apparel Manufacturers Association to the American Wind Energy Association, there are thousands of trade associations in the United States alone and thousands more in countries throughout the developed world (Aldrich & Staber, 1988; Bailey & Rupp, 2006; Collins & Roper, 2005; Lane & Bachmann, 1997). Yet “[v]ery little academic research has focused on trade associations” (Barringer & Harrison, 2000, p. 393). As a result, we have little systematic understanding of what these ubiquitous organizations actually do.

The lack of research on trade associations is lamentable. These organizations can amass budgets of millions of dollars and employ staffs of hundreds, and their lobbying activities can significantly influence the competitive environment, shaping regulations in ways that affect industry survival rates, growth, and profitability (Hirsch, 1975; Ingram & Inman, 1996; Miles, 1982). Many trade associations not only influence government regulation but even assume the role of government through industry self-regulation (Barnett & King, 2008; Gupta & Lad, 1983; Lenox & Nash, 2003; Maitland, 1985). Thus, if we are to understand the competitive environment, we cannot ignore the role of trade associations.

In this article, I examine the conditions that drive trade associations to action. Trade associations are organizations through which a group of interdependent firms, typically in the same industry, pool their resources and coordinate their efforts so that they may “speak with one voice”¹ on matters of shared interest. However, member firms are typically rivals and so they may not always share the same interests. Some firms may seek to direct the amassed resources of their trade associations in ways that maximize their individual interests and not necessarily the memberships’ shared interests. Empirically discerning what actually drives trade associations to “speak up”—be it the interests of a few powerful firms or the shared interests of the industry—is a critical step in better understanding these ubiquitous and influential organizations.

I conduct a longitudinal study of the activity of 148 major U.S. industry trade associations. Findings suggest that industry-wide downturns may not drive trade association activity. Rather, it is downturns in the performance of the dominant firms within an industry that are strongly associated with increased trade association activity. These findings thus imply that when trade associations speak with “one voice,” the most powerful firms may be writing the script.

What Do We Know About Trade Associations?

Trade associations are commonly defined as “organizations created to represent business interests within specific domains, mobilizing firms within their domain so that collective action can be taken on common problems” (Aldrich & Staber, 1988, p. 111). There are several thousand trade associations within the United States that represent the interests of almost every industry and thousands more across the industrialized world. Many are highly active, with large staffs, a wide range of active committees, and generous budgets. For example, the American Chemistry Council,² the primary trade association for nearly 200 chemical manufacturers, has a paid staff of approximately 300 personnel overseeing the association and its more than one dozen major subgroups, operating on an annual budget of around US$110 million (Swartout, 2008).

Not only the level but also the type of activity varies across trade associations. Hemphill (1992, p. 916) described the range of trade association activities as “data collection, educational programs, facilitating technical standards and specifications, insurance programs, legal assistance and government relations.” Oliver (1990, p. 249) listed trade association activities to include lobbying state regulators, promoting collective good through events such as trade shows, seeking to obtain economic advantages, reducing legislative uncertainty through activities such as product standardization, and enhancing members’ image. Staber and Aldrich (1983) distinguished four trade association “activity areas: commercial, public, political, and solidaristic” (p. 168). These encompass the organization of trade shows and conventions, product standardization, and the collection of industry statistics (commercial); compilation of statistics for regulatory compliance (public); lobbying the government (political); and building consensus and shared purpose within the industry (solidaristic).

Others have made a blunter distinction, wherein trade associations serve as the bodies through which firms pursue collective strategy in the political (Hillman & Hitt, 1999) or nonmarket (Baron, 1995) environments. This is an important distinction to make because colluding in market activities constitutes a violation of antitrust laws.³ Thus, legally, trade associations operate entirely within the nonmarket environment, which consists of “the social, political, and legal arrangements that structure interactions among companies and their public” (Baron, 1995, p. 73).

Though they cannot coordinate market activities, trade associations can enhance the performance and survival rates of their member firms by exerting influence on nonmarket forces. For example, the coordinated might of the pharmaceutical industry enabled it to co-opt critical resource holders and thereby gain higher average profits and survival rates than the recording industry, despite the similar structural characteristics of these two industries (Hirsch, 1975). The Tobacco Institute protected cigarette manufacturers from a variety of lawsuits and regulations that likely would have resulted in even greater harm to the industry (Miles, 1982). The coordinated effort of hoteliers around Niagara Falls led to regulation that overcame common problems that threatened their growth and prosperity (Ingram & Inman, 1996). Industry-wide cooperation through trade associations is particularly critical to recovery from crises that face entire industries (Barnett, 2006a, 2006b). By presenting a united front and speaking with a unified voice, firms may collectively “improve the legislative, regulatory, market, and public interest climate for the industry” (Chemical Manufacturers Association Report, 1993) and so help to realign their industry with the demands of its environment.

Yet despite their prevalence and potential impact, trade associations have received relatively little attention from organizational scholars (Barnett, Mischke, & Ocasio, 2000; Barringer & Harrison, 2000). The studies mentioned above focused on the effects that trade associations can have on parties external to an industry. The few other studies of trade associations have focused on population dynamics. Howard Aldrich, Udo Staber, and colleagues put forth several articles that explored variations in the rates of founding, transformation, and death of trade associations (Aldrich & Staber, 1988; Aldrich, Staber, Zimmer, & Beggs, 1990; Aldrich, Zimmer, Staber, & Beggs, 1994; Staber & Aldrich, 1983). They found that trade associations are very robust organizational forms due to their minimalist nature. Minimalist organizations are a special class of organizations that require few resources for founding and sustenance (Halliday, Powell, & Granfors, 1987). Trade associations require little overhead to initiate and, if necessary, can survive for extended periods of time with relatively little financial support. Members may withhold resources from a trade association in difficult times without causing the trade association to collapse. Thus, trade associations are relatively easy to found and likely to survive once founded.

Because of the minimalist nature of trade associations, population studies offer limited insight into key dynamics of these important organizations. Population studies are not sensitive to variation in organizational activity short of the disbanding of an existing organization or the establishment of a new one. An industry’s trade association is often founded early in the industry’s life because such collective action aids in gaining legitimacy for new types of organizations (Aldrich & Fiol, 1994; DiMaggio, 1988; Oliver, 1990). Yet once legitimacy is established, and so this shared problem is resolved, trade associations rarely disband. Rather, these robust organizations tend to survive for long periods of time. For example, though its name has changed several times, the American Chemistry Council has represented the evolving collective interests of chemical manufacturers since 1872. Thus, when an established industry faces a shared problem, its members are more likely to address this problem within the framework of an existing trade association than to create an entirely new one. As Aldrich and Staber (1988) acknowledged, population-level studies are not sensitive to this activity: “New positions might have been created, new committees formed, dues raised, lobbyists hired, and other actions taken . . . Such developments would not show up at the level of the analysis reported here” (p. 125).

What Do We Need To Know About Trade Associations?

To better understand the nature of trade associations, one must look beyond a few blunt milestones such as creation, reorganization, and eventual abandonment. Much happens in the intervening years. Some trade associations dominate their industries’ affairs, controlling large budgets and employing large staffs; others assume a minimalist posture and play a largely ceremonial role. Within a given trade association, activity waxes and wanes over time. Depending on the level of activity, trade associations can shape the competitive landscape, or they can be relatively inert. We need to look inside this black box and assess what drives variation in a trade association’s activity throughout its typically long life.

Collective Responses to Shared Problems

A trade association draws resources from its member firms. Firms often incur significant financial and human resource commitments as a condition of participation. For example, American Chemistry Council member firms pay annual dues ranging well into the hundreds of thousands of dollars.⁴ Furthermore, they provide “sweat equity” in the form of skilled personnel who participate in the thousands of meetings held at the Council’s headquarters each year (Rees, 1997). Yet membership and, accordingly, the associated resource commitments are voluntary. When will firms be willing to commit resources to their trade associations?

Profit-seeking firms, of course, prefer to limit their expenses, all else equal. In the case of trade association spending, this tendency would seem to be exacerbated because firms may engage in free riding. The benefits of trade association activity, such as an improved political and regulatory environment and a better industry reputation, are nonexcludable. When a trade association’s efforts leads to a reduction in the regulatory burden on an entire industry, all the firms in the industry benefit, including those who did not contribute to the trade association. Thus, as Olson’s (1965) influential work argued, the logic of collective action dictates that so long as participation in an initiative is voluntary and the fruits of the effort are nonexcludable, the rational choice is to free ride. In aggregate, this suggests that the “tragedy of the commons” (Hardin, 1968) prevails in these situations, producing not a voluntary organization that resolves shared problems but inaction that exacerbates these problems.

Despite the logic of collective action, a burgeoning body of research has shown that the tragedy of the commons is often averted in practice. Ostrom (1990) demonstrated that individuals did not succumb to the temptation to free ride under a number of circumstances relating to the preservation of shared natural resources. For example, individuals voluntarily created and participated in organizations that governed the use of fisheries and water sources. Others have extended these ideas on cooperation in the protection of natural resources to industrial settings. Rees (1994) found that the nuclear power industry, facing intense public criticism and rigid government oversight following a nuclear accident at Three Mile Island in 1979, quickly banded together through a trade association, the Institute of Nuclear Power Operations, to increase the industry’s prospects for survival. Barnett and King (2008) found that when the chemical industry was under threat following a deadly chemical spill, chemical firms came together to create a new industry self-regulatory program that significantly increased their voluntary commitments to their trade association.

In light of this growing body of research demonstrating that, when confronted by shared problems, firms can avert the tragedy of the commons and come together to create new institutional structures, I hypothesize that firms will increase their activity within existing trade associations when confronted by shared problems. The existing relationships between firms forged over time within a trade association facilitates mobilization by creating a common culture that unifies members (Abrahamson & Fombrun, 1994) and engenders peer pressure to cooperate (Gunningham, 1995), thereby helping overcome the collective action problem (Olson, 1965). In short, it is easier to respond to a shared problem through the action of an existing trade association than it is to create a new organization, and so if firms can create new organizations to manage shared problems, it seems only logical that they can take the less burdensome action of expanding activity within existing organizations to manage shared problems.

Hypothesis 1: A shared problem facing the members of an industry precipitates an increase in the activity of that industry’s trade association.

Collective Responses to the Problems of a Few

Though firms may overcome the collective action problem and voluntarily contribute to trade association efforts to manage shared problems facing their industry, how do they recognize a shared problem? The problems faced by a firm are often ambiguous (March & Olsen, 1976). The shared problems faced by an entire industry are yet more ambiguous. For example, how does a firm surmise that its industry’s legitimacy is under threat or that the regulatory climate has become problematic?

Hoffman and Ocasio (2001) found that the objective characteristics of a critical event confronting an industry often had little to do with the likelihood that the industry would attend to the event. Lacking objective guidelines on what constitutes a shared problem, firms may rely on social cues (DiMaggio & Powell, 1983). Firms tend to follow the actions of market leaders, presuming that if dominant firms are taking certain actions, then these actions must be appropriate (Ferrier, Smith, & Grimm, 1999). Thus, firms may be alerted to an issue and gauge its importance by the involvement of dominant firms.

In fact, without the leadership of large firms, the collective action problem inherent in trade association activity might never be surmounted (Olson, 1965). Large firms often put forth the initial effort and resources essential for beginning new collective endeavors (Hoffman, 1997; Rees, 1994). Moreover, the involvement of dominant firms increases the perceived feasibility of resolution, thereby increasing the likelihood that firms will participate and a coalition will arise. Faced with an array of potential issues vying for their limited attention, firms tend to participate in those issues that they perceive as most likely to be resolved because involvement with failed issues provides no rewards to offset the costs of participation (Bacharach & Lawler, 1981; Dutton & Webster, 1988).

The associated “follow the leader” behavior provides dominant firms the opportunity to set the agendas of their trade associations. Participation in a collaborative group such as a trade association is a mixed motive game involving simultaneous incentives to compete and cooperate (Phillips, Lawrence, & Hardy, 2000). Although member firms may share some common interests, they are ultimately rivals. There is no reason to believe that rivalry is set aside at the doorstep of a trade association. In fact, it seems more reasonable to believe that those firms with the power to sway trade association agendas actively seek to do so in their favor. Enterprising individuals and organizations have been shown to define, create, and manipulate issues to spur collective action toward some desired end (Davis & Thompson, 1994; McCarthy & Zald, 1977). Dominant firms in an industry are in a particularly opportune setting to define, create, and manipulate the problems that their trade association pursues.

The strategic intent behind some agenda setting efforts may be hidden. For example, a firm may champion a trade association program that mandates improvements in the social or environmental performance of an industry under the premise that the increased standards will improve the industry’s standing with regulators, customers, and other stakeholder groups. But if the championing firm can abide by these higher industry standards at a lower cost or in a superior way than can its rivals, then it gains competitive advantage. Large firms can have a strong hand in establishing these sorts of self-regulatory programs and so can shape these intraindustry policies in ways that play to their scale advantage. For example, Gunningham (1995) discussed the role of Dow Chemical in guiding the development of the chemical industry’s self-regulatory program, Responsible Care, and how Dow was able to benefit from this program more easily than were smaller firms who could not spread their costs over such a large asset base. A variety of studies have documented that firms may strategically pursue changes in government policy and regulation that actually increase their regulatory burden as a means of raising rivals’ costs even more and deterring new entrance (McWilliams, Van Fleet, & Cory, 2002; Salop & Scheffman, 1983; Williamson, 1968).

It is logical for large firms to initiate collective action on those issues for which they may receive private benefits more than ample to offset their costs of participation (Olson, 1965). But this logic does not suggest that smaller members of a trade association will knowingly subjugate their interests and willingly join efforts that disproportionately advantage their rivals. Especially in light of antitrust laws that limit the scope of coordination among rivals, it is not feasible for the dominant firms in an industry to force their trade association to pursue ventures that clearly disadvantage a substantial portion of the membership, no matter their relative power. However, many of the issues that trade associations face have subjective goals and largely intangible or distant measures of progress (e.g., choosing an ad campaign to improve public opinion, engaging a lobbyist to enhance the regulatory climate). Moreover, dominant firms hold considerable sway over smaller firms. They may use peer pressure or the threat of exclusion from business subcultures to forge cooperation (Galaskiewicz, 1985; Useem, 1984). This makes trade associations a fertile ground for enterprising firms to put forth their favored projects. In fact, a trade association provides the opportunity to engage in these practices more efficiently and effectively. By engaging a trade association instead of going it alone, a firm spreads the costs of its political and regulatory influence efforts.

In sum, the nature of the shared problems facing an industry is often subjective, which leaves the agenda of a trade association open to manipulation. Dominant firms have both the incentive and the ability to take the lead in mobilizing trade association activity. Other firms take their cue from dominant firms in deciding what constitutes a shared problem and whether or not collective action is likely to resolve it. Thus, dominant firms may seize the opportunity to mobilize trade associations to action to solve their particular problems rather than the problems of the industry as a whole. Therefore, I hypothesize the following:

Hypothesis 2: A problem facing the dominant members of an industry precipitates an increase in the activity of that industry’s trade association.

Data and Method

The lack of prior empirical research on activity within trade associations over time may be due, in part, to the difficulty of finding useful longitudinal data. The Encyclopedia of Associations (Swartout, 2008) is the primary source of data on trade associations. It annually reports activity figures such as membership, budget, and staffing for all trade associations. However, the Encyclopedia of Associations relies on self-reporting. Whether out of desire to disguise true spending or lack of desire to report annually, many trade associations report no figures, similar figures, or very round figures each year. Thus, it serves as a poor source of analysis of variability in spending.

Given these limitations, I turned instead to a source of compulsory reporting. Trade associations, as 501(c) nonprofit organizations, are required to file the Internal Revenue Service Form 990 every year. The Form 990 requires detailed reporting of a trade association’s income and expenses. A check of this data, made available by The Urban Institute,⁵ revealed that, indeed, trade association activity is not stagnant. For example, although the Encyclopedia of Associations reported an unchanging budget of US$3 million for the National Petroleum Refiners Association from 1990 to 2000, Form 990 data showed that this trade association’s total expenses ranged from US$3,610,483 to US$7,266,164, its gross income varied from US$2,969,870 to US$5,091,364, and its gross receipts changed from US$4,825,223 to US$7,838,843 over this same period. Thus, the Form 990 data provide several precise measures of a trade association’s annual activity.

Though the Encyclopedia of Associations data are of limited use in assessing variability in a trade association’s activity over time, they help in choosing a sample from among the large population of trade associations. The Encyclopedia of Associations lists more than 22,000 associations. From this list, I selected a sample of 148 major industry trade associations based on several criteria. The electronic version of the Encyclopedia of Associations, “Associations Unlimited,” provides several options for selecting a sample. I chose only those trade associations that were national in scope so as to avoid double-counting regional associations. Furthermore, I limited the sample to mature trade associations to factor out volatility in activity based on early industry growth. To do this, I included only those trade associations founded in 1980 or earlier—a full decade prior to the beginning of the study period. In addition, I excluded any trade association with more than 1,000 members because such large associations were likely to be professional associations representing the interests of individuals (e.g., lawyers, doctors, actors), not firms. Finally, I restricted the sample to those associations with significant annual budgets, defined as greater than US$1 million, to increase the likelihood that each trade association had more than just a symbolic role in furthering the interests of its members.

This search produced 255 trade associations. I then reviewed on-line documentation for each of these associations to screen out those trade associations that met the above search criteria but did not actually represent the interests of for-profit member firms within a particular industry. For example, the search returned trade associations such as the International Cotton Advisory Committee (countries, not firms as members), the Society of Professional Benefit Administrators (individuals, not firms as members), and the International Trade Council (peak or meta-association serving no single industry). Moreover, I excluded those associations that were not fully discretionary but were partially or fully overseen by a government agency (e.g., U.S. Wheat Associates). Finally, I excluded those trade associations for which I could not find a valid employer identification number (EIN), which serves as a link to the Form 990 database, as well as those trade associations that did not have Form 990 data for at least 8 years of the total 11-year period of this study (1990 to 2000). This resulted in a final sample of 148 major and mature industry trade associations and 1,589 trade association / year observations. The appendix provides a list of these trade associations along with some descriptive characteristics.

For the dependent variable in this study, I chose “annual total expenditures” (totexp). The Form 990 data provided annual figures for a variety of measures of trade association activity, such as the total dues paid, gross receipts, and gross income. I chose total expenditures because a trade association may accumulate assets and so has discretion as to when it may choose to add to or deplete these accumulated assets. Thus, a measure of total expenditures in a given year provides a more direct assessment of activity than does membership dues, gross receipts, or gross income. In other words, totexp assumes that trade associations will “put their money where their mouth is” and so will reveal their true agendas in the ways that they expend their resources over time.

There are two independent variables in this study—one measuring the annual performance of a given industry and the other measuring the annual performance of the dominant firms in that industry. Associations Unlimited provides one or more four-digit standard industrial classification (SIC) code for most trade associations. To augment this data where no SIC code was listed, to judge a primary SIC code where more than one SIC code was listed, and more generally, to serve as an additional check on what I have described above as often questionable data, an independent coder and I examined on-line documentation for each of the 148 trade associations. We independently determined the primary four-digit SIC code for each trade association. In any cases in which we had conflicting judgments, we discussed the cases and then came to a final mutual decision. In those cases in which our final choice of SIC conflicted with that of Associations Unlimited, we deferred to our judgment.

I measure the annual financial performance of a given industry with the variable “% profit for industry.” I computed % profit for industry as the total value of shipments for an industry in a particular year, less labor, material, and energy costs, all divided by total value of shipments. I obtained these figures from the U.S. Census Bureau’s Annual Survey of Manufacturers (ASM). Because of a change of classification system in 1997 (from SIC to NAICS), I could not obtain this measure after 1997. Thus, though the dependent variable spans 1990 to 2000, the time period over which the hypotheses can be tested is effectively restricted to the years 1990 to 1997.

To measure the annual financial performance of the dominant firms in a given industry (a test of Hypothesis 2), I turned to the Combined Compustat/CRSP Business Segment Database. ASM data are not available at the firm level. I computed the variable “% profit for top4” as the net operating profit (or loss) for the four largest firms (in terms of sales) in an industry as a percentage of that industry’s total assets. Compustat computes operating profit as operating revenue minus operating expense.

In addition to the above variables, I assembled several control variables. Trade associations with more members, all else equal, should have greater expenditures. Thus, I control for size through the variable “TA members.” Industries with high market concentration, wherein the bulk of market share is held by one or a few large firms, should have an easier time mobilizing for collective action than will industries with low market concentration, populated by a broad swath of small firms. I control for market concentration with the variable “HH index,” which is the Herfindahl–Hirschman index for each industry. The HH index takes account of the relative size and distribution of firms in an industry in forming an overarching measure of concentration. As noted above in my explanation of the chosen dependent variable, trade associations have discretion. I control for this discretion with the variables “age of TA” and “% rcpts not from dues.” As a trade association ages, it becomes more institutionalized and thus potentially more inertial. It may be difficult to disband the myriad projects and committees that arise over time. In addition, the more of a trade association’s receipts that are garnered through sources other than direct member dues, such as outside consulting, the more discretion the trade association maintains.

A major expense of trade associations is political lobbying. Therefore, I include a control for whether or not a trade association is headquartered in the Washington, D.C., area (DC office), as such positioning may be indicative of higher lobbying costs.⁶ Finally, to control for any effects on the dependent variable caused by variation in the overall economy over time, I include “year dummies.”

Table 1 provides descriptive statistics and correlation coefficients for these variables. Many of the correlations are significant. Therefore, I calculated variance inflation factors (VIFs) for each of the independent variables. These VIFs are well within the acceptable range, varying from 1.10 to 1.26, and so multicollinearity does not appear to be a problem.

Table 1.

Descriptive Statistics and Correlation Coefficients.

Variable	1	2	3	4	5	6	7	8
1. ln (totexp)
2. % profit for industry	−.028
3. % profit for top4	−.194**	.348**
4. ln (TA members)	.154**	.035	.048
5. ln (HH index)	.208**	−.104	.067	−.204**
6. Age of TA	.334**	.082	.024**	.144**	.162**
7. % rcpts not from dues	0*	−.139	.133**	.338**	.014	−.064
8. DC office	.368**	.04	.036	.04	.169**	.226**	−.12**
n	1,583	624	1,246	1,589	970	1,589	1,567	1,589
M	14.8	0.3	0.08	5.3	6	60	0.44	0.66
SD	1.2	0.13	0.07	1.2	1.2	26	0.26	0.48
Minimum	11	-0.17	−0.15	1.8	2.6	20	0	0
Maximum	18.6	0.65	0.46	7.5	7.9	128	0.99	1

p < .05. **p < .01.

Results

Table 2 provides the results for several regression models predicting totexp. The odd-numbered models show the results of regressions that did not include % profit for top4, whereas the even-numbered models include this variable. Adding % profit for top4 into the regression separately allows one to more clearly distinguish the effects on the dependent variable attributed to the performance of the dominant firms (Hypothesis 2) from that of the overall industry (Hypothesis 1).

Table 2.

Regression Results Predicting Trade Association Total Expenditures.

	OLS		Robust		Cluster		Between Effects		Fixed Effects
Variable	1	2	3	4	5	6	7	8	9	10
Constant	12.1***	12.1***	12.3***	12.1***	12.4***	12.1***	9.8***	8.8**	14.5***	14.6***
% profit for industry	-.306	.374	.005	.731*	-.306	-.374	-.255	.569	-.142	-.37
% profit for top4		-3.98***		-3.82***		-3.98***		-5.98***		-0.037
ln (TA members)	.108**	.147***	.129***	.148***	.108	.147	.091	.158	D	D
ln (HH index)	.143***	.182***	.158***	.194***	.143**	.182**	.145	.183	D	D
Age of TA	.01***	.01***	.007***	.007***	.01*	.01*	.011*	.01**	D	D
% rcpts not from dues	-.03	.205	-.184	.135	-.03	.205	-.097	.122	.71***	.801***
DC office	.664***	.726***	.624***	.699***	.664***	.726***	.604**	.716***	D	D
Year dummies	Incl.	Incl.	Incl.	Incl.	Incl.	Incl.	Incl.	Incl.	Incl.	Incl
No. of observations	617	489	617	489	617	489	617	489	617	489
No. of clusters/groups					90	78	90	78	90	78
Adjusted R2	.20	.28			.22	.30	.26a	.39a	.33b	.39b
F			11.3***	12.3***

Note: OLS = ordinary least squares; D = dropped from model.

Between groups (trade associations) R².

Within groups (trade associations) R².

p < .1. **p < .05. ***p < .01.

Model 1 is an ordinary least squares regression model. The results of Model 1 show that % profit for industry is not a significant predictor of totexp, thus providing no support for Hypothesis 1. When % profit for top4 was added to the regression in Model 2, % profit for industry remained insignificant, but % profit for top4 was significant and negative at the p < .01 level. These results indicate support for Hypothesis 2, wherein trade association expenditures are high when the profitability of dominant firms in an industry is low.

Across both models, several control variables produced significant results. TA members was positively related to totexp, suggesting that, as expected, trade associations with more members tend to have higher expenditures. HH index was also positive and significant, indicating that industries with greater concentration have higher trade association spending. Age of TA was also positive and significant, pointing to a somewhat slight tendency for older trade associations to have higher expenditures. Moreover, trade associations headquartered in the Washington, D.C., area, as expected, tended to outspend those headquartered elsewhere. The percentage of receipts of a trade association generated from sources other than member dues appeared to have no significant effect on trade association expenditures.

Models 3 and 4 show the output of a robust regression (rreg method in Stata). Robust regression limits the effects of outliers through weighting (Rousseeuw & Leroy, 1987). I ran robust regression to ensure that none of the larger trade associations had undue influence on the results. Again, the results support Hypothesis 2 and provide no support for Hypothesis 1. In fact, Model 4 shows some slight support for the notion that trade association activity is low when the profitability of the overall industry is low. This may be due to the difficulty of garnering resources from member firms when they face resource constraints. All control variables follow a very similar pattern to that found in Models 1 and 2.

Across this panel, all observations are not independent. The total expenditures of a given trade association are generally not independent of its expenditures in prior years. To account for this, in Models 5 and 6, I ran a robust cluster model. This model relaxes the assumption of independent observations within clusters; in this case, each cluster represents a single trade association. The results again appear to support Hypothesis 2 and provide results quite similar to the prior models, though several of the control variable effects become weaker.

As a stronger control for the lack of independence of observations within a trade association, Models 7 and 8 show the results of a between-effects regression model. This model effectively treats each trade association as a single observation and instead focuses on measuring variation across trade associations. The results are quite consistent with prior models, again lending support to Hypothesis 2, but further highlighting a constraint inherent in this study: The effects found within this panel are cross-sectional, not longitudinal.

Because of data constraints, the portion of the panel included in these regressions encompasses only the period from 1990 to 1997. The total expenditures within a given trade association often do not vary greatly over this time period. As a result, the data reveal little about differences in performance within a given trade association over the time period and instead serve as a test of differences across trade associations. Models 9 and 10, which are fixed effects regressions, confirm this. The fixed effects model factors out the average total expenditures for each trade association and seeks to explain differences within each trade association that deviate from this average factor or fixed effect. Both independent variables are insignificant in the fixed effects models. Moreover, the year dummies, though not listed independently in Table 2, become highly significant (p < .01) in these models yet were generally not significant in the prior models.

The variables were assembled from differing sources, so not all variables were available for all trade associations for all years, leading to a significant decrease in the number of trade association / year observations included in the regression models. For example, some of the variables were available only for trade associations with manufacturing SIC codes, from 1990 to 1997. Other variables were available for all years and nonmanufacturing SIC codes but often not all of the same SIC codes as the other variables. Thus, there are 617 observations in the odd-numbered models and 489 in the even-numbered models. Because these two samples may have differing characteristics, I also ran each odd-numbered model on only those 489 observations in the even-numbered model sample. The results were nearly identical and produced identical directionality.

Some of the effects of the independent variables on the dependent variable may be lagged, as it may take some time to mobilize the trade association to increase spending in response to a decline in performance. I ran all 10 models with 1- and 2-year lags. This lagging further reduced the data set, providing somewhat weaker fit but overall similar results in terms of significance and directionality. These results are again consistent with the problem of limited variability in spending across years producing effectively a cross-sectional study.

Discussion

The lack of research on trade associations has left a large and important part of the competitive landscape uncharted. The findings of this study are limited in significant ways, so they cannot provide a detailed map of the uncharted territory of trade associations, but they do suggest some basic points of interest that are worthy of more detailed mapping. This study was unable to explain variations in the activity of a given trade association over time. Nonetheless, in a robust manner and controlling for a variety of variables, this study found that across trade associations, low performance of their primary industry’s four largest firms was associated with high trade association expenditures, yet low performance of their primary industry was not associated with high trade association expenditures. This suggests that trade associations may speak for and serve the interests of the dominant firms in an industry more so than the interests of the industry as a whole.

Some of the findings across the models offer more nuance. Model 4 found that the performance of the industry was positively, though weakly, associated with the total expenditures of its trade association. Though inferences must be made with caution given the limits of the data, this finding suggests that not only may the industry’s voice be primarily that of the dominant firms, but these dominant firms may have an easier time pushing their agendas when the members firms have slack resources, as contrasted with the difficulty they would face in garnering additional funds to pursue their agendas during lean times.

The findings of the fixed effects analyses offer a bit more insight. Models 9 and 10 show a strong positive relationship between the percentage of a trade association’s receipts derived from sources other than membership dues and trade association spending. This suggests that as trade associations secure more funding independent of direct dues payments, they tend to make use of this discretion in spending rather than by contributing to reserves. The average nondues income in the sample is 44%, and it can rise to as high as 99%. Thus, it appears that trade associations are fairly independent bodies, able to survive, and spend, in the near term without necessarily having to directly raise member dues. Instead, a significant portion of receipts may come from interest on investments or the sale of assets, or from trade association activities such as program service revenues and meetings and seminars. With such a high level of financial discretion, it seems all the more important to understand how trade associations choose to spend their funds.

Trade associations occupy a tricky legal space that sometimes borders on violation of antitrust laws (Luria, 2005). Though trade associations explicitly involve coordination among rivals, trade associations are allowed to exist under the premise that they serve as a means of resolving shared problems in a way that advances competition while avoiding collusion. If, indeed, trade associations tend to be tools of domination by powerful firms, their antitrust exemption becomes questionable. The findings in this article are not adequate to draw such a conclusion, but they do point to the need for further investigation of just how trade associations work.

For firms that do not dominate their industries—that is, most firms—these findings should give pause for reflection on the benefits they receive relative to the contributions they make to their trade associations. Paying trade association dues may be more or less a rote exercise for most firms, and managers may have no more an expectation of measurable financial returns from these investments than they have from, say, sponsorship of a local little league team. However, when member firms are called on to increase their involvement or support particular trade association initiatives, they should not simply assume that these efforts are in their best interests and so worth pursuing. Instead, they should give consideration to the possibility that they may be using their limited resources in a way that contributes more to the advancement of their rivals’ interests than to their industry’s shared interests.

The management research on trade associations is limited, but it does recognize that these organizations face an ongoing tension associated with size (Aldrich, 1999; Schmitter & Streeck, 1981; Van Waarden, 1992). Each trade association, as an organization pursuing its own interests, wishes to gain membership and grow larger. Member firms can benefit from increased size because it can bring with it increased influence. However, as a trade association gets larger and its membership grows more diverse, it becomes less able to fully satisfy the unique needs of its individual members, and so this leads to pressure to splinter into subgroups. The findings of this study cannot conclude when a firm no longer fits with its trade association, but it does provide some suggestion that smaller firms may be remaining in large associations longer than they should and might better be served by splintering into smaller specialized trade associations with less heterogeneity in size and power. A smaller trade association means less influence, but it may be that this lessened influence is of greater benefit to its members than is being part of a larger organization that predominantly serves the interests of larger rivals.

Overall, this study helps bring the field a little closer to understanding the drivers of trade association activity, but obviously much more needs to be done. For one, future studies would benefit greatly from lengthier windows. Prior studies in the ecological tradition had very lengthy windows but did not focus on variations in activity within trade associations. The combination of long windows and performance data would be ideal. Trade association activity appears to be fairly constant in the short run. Longer time periods are necessary to evince any measurable disruptions. However, given the underdeveloped state of this literature, it may be necessary to first proceed with smaller scale studies. Qualitative studies may be necessary to observe and understand the actual mechanisms. Although motives can be inferred from spending patterns and other such outcome measures over time, it would be particularly insightful to directly observe trade association activity through field studies. Such work would be laborious, of course, and it could be difficult to gain access to trade associations, but where feasible, such studies offer great promise of rich insight into the workings of these important organizations.

Future studies should also look at the ways in which factors other than changes in profit affect changes in trade association activity. I operationalized a shared problem as a decline in the industry’s profitability. Collective action can be difficult to organize (Olson, 1965) especially in the face of ill-defined threats. A decline in profitability would seem to be a tangible and significant threat and so a conservative measure of what might spur firms to take collective action through their trade associations. However, trade associations may not be spurred to action when profit declines if they perceive the decline in profit as relating to factors outside of their control. The inclusion of year effects in this study helped alleviate some concern that exogenous macroeconomic factors such as a general market downturn in any given year might be responsible for trade association activity. However, this study could not determine whether firms perceived any change in industry profitability as something they could influence through trade association activity. Moreover, it is possible that firms may perceive other actions as shared threats, to cope with through trade association activity, before they cause any decline in industry profitability. For example, firms may recognize negative press coverage, looming lawsuits, or a general decline in reputation scores as signaling problems and take action in advance. Future studies should further explore how firms come to recognize shared threats and just how serious the threat must be perceived to be before it spurs collective action (cf. Hoffman & Ocasio, 2001).

In this study, I stepped down from the population level of analysis at which the few prior studies in the organizations literature resided (e.g., Aldrich & Staber, 1988) to look inside trade associations. Nonetheless, the level of insight I was able to garner with the data available was still too abstract to enable strong inference about the inner workings of trade associations. To develop richer insights, future studies will need to gather data that likely will only be feasible to collect through smaller sample sizes. For example, with 148 trade associations, each with membership bases ranging into the hundreds and spanning many years, I was unable to verify that the largest firms in each industry were always members of the trade association affiliated with that industry in this study. Firms may come and go from a trade association over time for any of a number of reasons. Thus, it is possible that even though a firm is a dominant firm in an industry, in any given year it may not be a member of that industry’s major trade association.

Trade associations generally do not provide direct access to their membership archives over time. In this study, I checked a sample of five trade associations to confirm that the current membership rosters contained the “top four” firms in their industry during the study period. Of the 20 firms, I could not account for 2. Because I do not have the membership rosters over time and instead am relying on a current membership roster, it is possible that these two firms were members during the study period, but I cannot say with certainty. However, Staber’s (1987) interviews of trade associations in the manufacturing sector add confidence that this is unlikely to be a significant problem:

Most associations are highly representative of the domain they cover, regardless of their size. Ninety-three per cent cover more than half of their members’ industry share, and 74 per cent represent more than three-quarters of their domain . . . all but one represent the four largest producers in their industry. (p. 284)

Another limitation of the large sample study I conducted is that I was unable to account for the performance of private firms. I relied on Compustat/CRSP for firm-level performance data. As a result, though I tested for whether or not trade association activity was reliant on changes in the performance of dominant firms, I was unable to contrast changes in large firm performance with changes in the profitability of smaller firms, or even many medium-sized firms in the same industry, since these tend not to be publicly listed. Smaller sample studies might be able to obtain useful performance measures for the entirety of their member firms and so contrast large firm effects with those of medium and small firms to better specify the true drivers of trade association activity.

Finally, it is important to note that the broad view of trade associations taken in this study did not allow me to determine the fit between the mission of any particular trade association and the performance of the particular four-digit SIC industry to which I assigned it. As noted previously, I did follow a systematic process for determining the appropriate industry for a given trade association. But this process cannot ensure that the firms within a given SIC rely on that particular trade association exclusively to manage their shared problems. Future smaller scale studies might be able to determine performance of the relevant population of a given trade association at a finer-grained level than that of the four-digit SIC code.

Conclusion

Because firms face many critical interdependencies, strategy researchers have broadly concluded that competition alone is insufficient for the success and survival of most firms. Environmental conditions often affect the likelihood of survival not of individual firms but of groups of firms and entire industries (Astley & Fombrun, 1983). Firms may find that their reputations and performance are intertwined such that the acts of one firm can harm similar other firms (Barnett & Hoffman, 2008; King, Lenox, & Barnett, 2002). To cope with these interdependencies, rivalrous firms often must cooperate (Barnett, 2006a). Especially, in recent years, as many industries have come under attack from increasingly powerful constituent groups (Davis & Thompson, 1994; Freeman, 1984) and government has relaxed many relevant antitrust concerns, firms have coordinated more of their efforts through trade associations. Many industry trade associations now go so far as to stringently regulate and standardize the behavior of member firms through industry self-regulation (Barnett & King, 2008; Prakash & Potoski, 2006). Thus, it is all the more important to understand what goes on within trade associations over time.

It is important to note that firms do not rely exclusively on trade associations to deal with problems in their competitive environments. “Communal strategies” (Barnett, 2006a) may be operationalized through venues other than industry trade associations or through multiple trade associations. In addition, when individual firms, especially dominant firms, experience difficult times, they may engage in a variety of individual strategies, such as lobbying and contributions to political action committees. The literature has tended to describe such activities in terms of a dichotomy—firms either act individually to pursue their own nonmarket or political activities, or they join collective efforts such as those of trade associations (see, for example, Hillman & Hitt, 1999). The costs of individually pursuing these activities are high and often may not be effective at favorably influencing the competitive environment. As Astley and Fombrun (1983) argued, “the very impotence of organizations acting in isolation merely elevates the importance of collective action” (p. 582). As a result, trade associations, wherein the costs are pooled and the influence enhanced, are often the primary means of influencing the competitive environment. The findings of this study suggest a sort of third “hybrid” form whereby large firms might use trade associations to pursue their individual goals, effectively leveraging the might of the trade association at a fractional cost of engaging in these efforts alone. Future studies should seek to parse out the degree to which they turn to each of these varying activities. The more refined such work becomes the closer we will come to achieving “the most basic operational requirement in the formulation and implementation of strategy,” which is to explain the degree to which rivalrous firms also choose to “take on responsibilities as members of a larger social entity” (Astley & Fombrun, 1983, p. 585).

Footnotes

Appendix

Trade Associations in Sample

Name of Trade Association	Founded	Classification	Standard Industrial Staff	Members
Adhesive and Sealant Council	1957	2891	10	130
Aerospace Industries Association of America	1919	3721	44	56
Air Conditioning and Refrigeration Institute (ARI)	1953	3585	45	207
Aluminum Association	1933	3334	30	54
American Architectural Manufacturers Association	1962	3272	14	280
American Association of Advertising Agencies	1917	7311	80	505
American Crop Protection Association	1933	5191	40	78
American Film Marketing Association	1980	7812	35	180
American Insurance Association	1964	6311	140	350
American Iron and Steel Institute	1908	3325	50	50
American Pet Products Manufacturers Association	1959	5199	14	600
American Petroleum Institute	1919	2911	270	400
American Short Line Railroad Association	1913	4011	9	775
American Warehousemen’s Association	1891	4225	10	550
American Waterways Operators	1944	4491	20	375
American Wind Energy Association	1974	3511	13	850
American Bakers Association	1897	2051	14	300
American Council of Life Insurance (ACLI)	1976	6311	184	427
American Feed Industry Association	1909	2048	17	675
American Financial Services Association (AFSA)	1916	6061	30	569
American Gas Association (AGA)	1918	5172	100	185
American Gear Manufacturers Association (AGMA)	1916	3462	13	410
American Wood Preservers Institute (AWPI)	1920	2491	7	120
Asphalt Institute (AI)	1919	2911	42	48
Association of International Automobile Manufacturers	1964	3711	11	17
Association of Oil Pipe Lines	1947	4612	5	56
Association of Rotational Molders	1976	3089	6	460
Association of Air Medical Services (AAMS)	1980	4522	5	400
Association of American Publishers (AAP)	1970	2731	23	310
Association of Directory Publishers (ADP)	1898	2741	6	274
Automotive Warehouse Distributors Association	1947	5013	23	335
Battery Council International	1924	3691	3	175
Bituminous Coal Operators’ Association (BCOA)	1950	1221	4	19
Bond Market Association	1977	6211	70	200
Chemical Manufacturers Association	1872	2819	300	195
Chlorine Institute	1924	2812	12	240
Chocolate Manufacturers Association of the USA	1923	2064	7	11
Cigar Association of America (CAA)	1937	2121	5	61
Council For Responsible Nutrition	1973	2834	11	100
Commercial Finance Association	1944	6282	15	290
Composite Panel Association	1960	2493	12	36
Composites Fabricators Association (CFA)	1979	2221	14	800
Compressed Gas Association (CGA)	1913	2813	16	230
Copper Development Association (CDA)	1963	3331	22	70
Cosmetic Toiletry and Fragrance Association	1894	2844	49	600
Council of Insurance Agents and Brokers	1913	6411	18	100
Direct Selling Association	1910	5963	22	200
Edison Electric Institute	1933	4911	260	240
Engine Manufacturers Association	1968	3519	5	29
Envelope Manufacturers Association of America	1933	2677	7	175
Equipment and Tool Institute	1947	3559	4	70
Equipment Leasing Association of America (ELA)	1961	7353	27	850
Flexible Packaging Association	1950	3089	20	150
Forging Industry Association (FIA)	1913	3312	13	230
Glass Packaging Institution	1945	3221	5	43
Grocery Manufacturers of America	1908	5141	40	135
Hardwood Plywood and Veneer Association	1921	2435	11	195
Health Industry Distributors Association (HIDA)	1902	5047	21	850
Health Industry Manufacturers Association	1974	3841	55	800
Health Insurance Association of America	1956	6321	100	290
Health Care Distribution Management Association	1876	5122	41	456
Hearing Industries Association (HIA)	1957	3842	4	32
INDA—Association of the Nonwoven Fabrics Industry	1968	2297	20	220
Independent Liquid Terminals Association	1974	4226	6	70
Industrial Fasteners Institute	1931	3452	8	150
Industrial Truck Association	1924	3537	5	145
Information Technology Industry Council	1916	3571	30	28
International Association of Plastics Distributors	1956	5162	7	450
International Mass Retail Association (IMRA)	1966	5399	25	750
International Sleep Products Association (ISPA)	1915	2515	18	750
International Jelly and Preserve Association	1945	2033	5	75
International Tape-Disc Association	1970	3651	9	450
Juvenile Products Manufacturers Association	1962	2512	10	400
Landscape Nursery Council (LANCO)	1952	0181	2	11
Lead Industries Association (LIA)	1928	1031	2	80
Leather Industries America	1917	3111	6	250
Luggage and Leather Goods Manufacturers of America	1938	3161	9	240
Magazine Publishers of America	1919	2721	37	200
Mailing and Fulfillment Service Association (MFSA)	1920	7331	11	720
Manufactured Housing Institute (MHI)	1936	2452	25	350
Material Handling Industry	1945	5084	30	700
Metal Powder Industries Federation	1944	3399	18	315
Motion Picture Association of America	1922	7812	120	8
Motor and Equipment Manufacturers Association	1904	3711	90	700
Motorcycle Industry Council (MIC)	1914	3751	20	310
Napa Valley Vintners	1943	2084	15	150
National Association of Hosiery Manufacturers, NC	1905	2251	9	425
National Association of Small Business Investment	1958	6159	6	300
National Concrete Masonry Association	1918	3271	29	500
National Confectioners Association	1884	2064	27	700
National Electrical Manufacturers Association	1926	3621	95	550
National Electronic Distributors Association	1937	5065	8	300
National Fluid Power Association	1953	3492	16	250
National Food Processors Association	1909	2099	185	500
National Insulation Association	1953	1742	9	700
National Investment Company Service Association	1962	6211	7	400
National Milk Producers Federation	1916	241	20	25
National Multi Housing Council	1978	6531	31	900
National Oilseed Processors Association	1929	2079	6	13
National Paint and Coatings Association	1933	2851	40	450
National Petroleum Refiners Association	1902	2911	31	480
National Precast Concrete Association	1965	3272	13	700
National Renderers Association	1933	2077	5	300
National Venture Capital Association	1973	6799	10	335
National Asphalt Pavement Association	1955	1611	25	750
National Association of Chemical Distributors (NACD)	1971	5169	7	265
National Association of Independent Insurers (NAII)	1945	6331	190	690
National Association of Recording Merchandisers	1958	5735	14	1,000
National Association of Water Companies (NAWC)	1895	4941	10	340
National Chicken Council (NCC)	1954	0251	12	225
National Coffee Association of USA (NCA)	1911	5149	5	185
National Coil Coating Association (NCCA)	1962	3479	6	167
National Soft Drink Association	1919	2086	36	865
National Wooden Pallet and Container Association	1947	2448	10	525
Nonprescription Drug Manufacturers Association	1881	2834	39	215
North American Insulation Manufacturers Association	1933	3296	8	13
Optical Laboratories Association	1894	3851	8	375
Outdoor Power Equipment Institution	1952	3524	8	78
Paperboard Packaging Council	1967	2652	8	65
Personal Communications Industry Association	1965	4812	75	550
Pharmaceutical Care Management Association	1975	6321	16	147
Physician Insurers Association of America	1977	6324	14	60
Portland Cement Association	1916	3241	90	50
Power Transmission Distributors Association	1960	3566	7	465
Precision Machined Products Association	1933	3451	16	550
Promotion Marketing Association of America	1911	8742	12	700
Real Estate Roundtable	1969	6531	11	200
Recording Industry Association of America	1952	3652	60	250
Recreation Vehicle Industry Association	1973	3711	56	520
Reinsurance Association of America	1969	6331	24	36
Rubber Manufacturers Association	1915	2822	22	97
Semiconductor Industry Association	1977	3674	12	40
Snack Food Association	1937	2096	13	700
Society of Independent Gasoline Marketers of America	1958	5172	10	350
Solar Energy Industries Association (SEIA)	1974	3433	4	500
Steel Tube Institute	1930	3312	3	75
Sugar Association	1949	2062	10	23
Synthetic Organic Chemical Manufacturers Association	1921	2869	45	300
The Sulphur Institute (TSI)	1960	2819	9	30
Toy Manufacturers America	1916	3944	24	300
Truck Renting and Leasing Association (TRALA)	1978	4212	7	700
Trucking Management	1963	4212	7	6
Uniform and Textile Service Association (UTSA)	1933	3582	15	200
Valve Manufacturers Association of America	1938	3491	7	120
Vinegar Institute	1967	2099	3	45
Wheat Foods Council (WFC)	1972	0111	3	50
Wine and Spirits Wholesalers of America (WSWA)	1943	5182	11	530
World Wide Pet Supply Association	1951	2047	6	550

Author’s Note

This article stems from work associated with my dissertation. My thanks to mydissertation committee members: Bill Starbuck, Andy King, and Brian Uzzi.

Notes

Author Biography

Michael L. Barnett is Professor 1 in the Management and Global Business Department and Vice Dean for Academic Programs at Rutgers Business School–Newark and New Brunswick. Mike’s research focuses on the firm-stakeholder interface. In particular, he studies how firms individually and collectively manage their relationships with stakeholders, and how their efforts at stakeholder management, through acts of corporate social responsibility and via communal institutions such as industry trade associations, influence their reputations and financial performance.

References

Abrahamson

Fombrun

(1994). Macrocultures: Determinants and consequences. Academy of Management Review, 19, 728-755.

Aldrich

(1999). Organizations evolving. Thousand Oaks, CA: Sage.

Aldrich

Fiol

C. M.

(1994). Fools rush in? The institutional context of industry creation. Academy of Management Review, 19, 645-671.

Aldrich

Staber

U. H.

(1988). Organizing business interests: Patterns of trade association foundings, transformations, and deaths. In Carroll

G. R.

(Ed.), Ecological models of organizations (pp. 111-126). Cambridge, MA: Ballinger.

Aldrich

Staber

U. H.

Zimmer

Beggs

(1990). Minimalism and organizational mortality: Patterns of disbanding among U.S. trade association, 1900-1983. In Singh

J. V.

(Ed.), Organizational evolution: New directions (pp. 21-52). Beverly Hills, CA: Sage.

Aldrich

Zimmer

Staber

U. H.

Beggs

(1994). Minimalism, mutualism, and maturity: The evolution of American trade associations in the 20th century. In Singh

Baum

(Eds.), The evolutionary dynamics of organizations (pp. 223-239). Oxford, UK: Oxford University Press.

Astley

Fombrun

(1983). Collective strategy: Social ecology of organizational environments. Academy of Management Review, 8, 576-587.

Bacharach

Lawler

(1981). Power and politics in organizations. San Francisco: Jossey Bass

Bailey

Rupp

(2006). The evolving role of trade associations in negotiated environment agreements: The case of United Kingdom climate change agreements. Business Strategy and the Environment, 15, 40-54.

10.

Barnett

(2006a). Finding a working balance between competitive and communal strategy. Journal of Management Studies, 43, 1753-1773.

11.

Barnett

(2006b). Waves of collectivizing: A dynamic model of competition and cooperation over the life of an industry. Corporate Reputation Review, 8, 272-292.

12.

Barnett

Hoffman

(2008). Beyond corporate reputation: Managing reputational interdependence. Corporate Reputation Review, 11, 1-9.

13.

Barnett

King

(2008). Good fences make good neighbors: An institutional explanation of industry self-regulation. Academy of Management Journal, 51, 1150-1170.

14.

Barnett

Mischke

Ocasio

(2000). The evolution of collective strategies among organizations. Organization Studies, 21, 325-354.

15.

Baron

(1995). The nonmarket strategy system. Sloan Management Review, 37, 73-85.

16.

Barringer

B. R.

Harrison

J. S.

(2000). Walking a tightrope: Creating value through interorganizational relationships. Journal of Management, 26, 367-403.

17.

Chemical Manufacturers Association Report. (1993). On the road to success (Responsible Care progress report). Washington, DC: Author.

18.

Collins

Roper

(2005). Strategic schizophrenia: The strategic use of trade associations in New Zealand. Journal of Communication Management, 9, 256-266.

19.

Davis

Thompson

(1994). A social movement perspective on corporate control. Administrative Science Quarterly, 39, 141-173.

20.

DiMaggio

(1988). Interest and agency in institutional theory. In Zucker

(Ed.), Institutional patterns and organizations (pp. 3-22). Cambridge, MA: Ballinger.

21.

DiMaggio

Powell

W. W.

(1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48, 147-160.

22.

Dutton

Webster

(1988). Patterns of interest around issues: The role of uncertainty and feasibility. Academy of Management Journal, 31, 663-675.

23.

Ferrier

Smith

Grimm

(1999). The role of competitive actions in market share erosion and industry dethronement: A study of industry leaders and challengers. Academy of Management Journal, 42, 372-388.

24.

Freeman

(1984). Strategic management: A stakeholder approach. Boston: Pitman.

25.

Galaskiewicz

(1985). Interorganizational relations. Annual Review of Sociology, 11, 281-304.

26.

Gunningham

(1995). Environment, self-regulation, and the chemical industry: Assessing Responsible Care. Law & Policy, 17, 59-109.

27.

Gupta

Lad

(1983). Industry self-regulation: An economic, organizational, and political analysis. Academy of Management Review, 8, 416-425.

28.

Halliday

Powell

Granfors

(1987). Minimalist organizations: Vital events in state bar associations, 1870-1930. American Sociological Review, 52, 456-471.

29.

Hardin

(1968). The tragedy of the commons. Science, 162, 1243-1248.

30.

Hemphill

(1992). Self-regulating industry behavior: Antitrust limitations and trade association codes of conduct. Journal of Business Ethics, 11, 915-920.

31.

Hillman

Hitt

(1999). Corporate political strategy formulation: A model of approach, participation, and strategy decisions. Academy of Management Review, 24, 825-842.

32.

Hirsch

(1975). Organizational effectiveness and the institutional environment. Administrative Science Quarterly, 20, 327-344.

33.

Hoffman

(1997). From heresy to dogma: An institutional history of corporate environmentalism. San Francisco: New Lexington Press.

34.

Ingram

Inman

(1996). Institutions, intergroup rivalry, and the evolution of hotel populations around Niagara Falls. Administrative Science Quarterly, 41, 629-658.

35.

Hoffman

Ocasio

(2001). Not all events are attended equally: Toward a middle-rangetheory of industry attention to external events. Organization Science, 12, 414-434.

36.

King

Lenox

Barnett

(2002). Strategic responses to the reputation commons problem. In Hoffman

Ventresca

(Eds.), Organizations, policy, and the natural environment: Institutional and strategic perspectives (pp. 393-406). Stanford, CA: Stanford University Press.

37.

Lane

Bachmann

(1997). Cooperation in inter-firm relations in Britain and Germany: The role of social institutions. British Journal of Sociology, 48, 226-254.

38.

Lenox

Nash

(2003). Industry self-regulation and adverse selection: A comparison across four trade association programs. Business Strategy and the Environment, 12, 343-356.

39.

Lenway

Rehbein

(1991). Leaders, followers, and free riders: An empirical test of variation in corporate political involvement. Academy of Management Journal, 34, 893-905.

40.

Luria

(2005). Trade associations face important patent/antitrust issues. Intellectual Property & Technology Law Journal, 17, 13-19.

41.

Maitland

(1985). The limits of business self-regulation. California Management Review, 27, 132-147.

42.

March

J. G.

Olsen

J. P.

(1976). Ambiguity and choice in organizations. Bergen, Norway: Universitetsforlaget.

43.

McCarthy

Zald

(1977). Resource mobilization and social movements: A partial theory. American Journal of Sociology, 82, 1212-1241.

44.

McWilliams

Van Fleet

Cory

(2002). Raising rivals’ costs through political strategy: An extension of resource-based theory. Journal of Management Studies, 39, 707-723.

45.

Miles

R. H.

(1982). Coffin nails and corporate strategies. Englewood Cliffs, NJ: Prentice-Hall.

46.

Oliver

(1990). Determinants of interorganizational relationships: Integrations and future directions. Academy of Management Review, 15, 241-265.

47.

Olson

(1965). The logic of collective action: Public goods and the theory of groups. Cambridge, MA: Harvard University Press.

48.

Ostrom

(1990). Governing the commons: The evolution of institutions for collective action. Cambridge, UK: Cambridge University Press.

49.

Phillips

Lawrence

Hardy

(2000). Inter-organizational collaboration and the dynamics of institutional fields. Journal of Management Studies, 37, 23-43.

50.

Prakash

Potoski

(2006). The voluntary environmentalists: Green clubs, ISO 14001, and voluntary environmental regulations. Cambridge, UK: Cambridge University Press.

51.

Rees

(1994). Hostages of each other: The transformation of nuclear safety since Three Mile Island. Chicago: University of Chicago Press.

52.

Rees

(1997). The development of communitarian regulation in the chemical industry. Law & Policy, 19, 477-528.

53.

Rousseeuw

P. J

Leroy

A. M.

(1987). Robust regression and outlier detection. New York: Wiley.

54.

Salop

Scheffman

(1983). Raising rivals’ costs. American Economic Review, 73, 267-271

55.

Schmitter

Streeck

(1981). The organization of business interests (IIM/LMP 81-13). Berlin, Germany: International Institute of Management, Labour Market Policy Group.

56.

Schuler

(1996). Corporate political strategy and foreign competition: The case of the steel industry. Academy of Management Journal, 39, 720-737.

57.

Staber

(1987). Corporatism and the governance structure of American trade associations. Political Studies, 35, 278-288.

58.

Staber

Aldrich

(1983). Trade association stability and public policy. In Hall

Quinn

(Eds.), Organizational theory and public policy (pp. 163-178). Beverly Hills, CA: Sage.

59.

Swartout

(Ed.). (2008). Encyclopedia of associations: National organizations of the U.S. (46th ed.). Detroit, MI: Gale Research Co.

60.

Useem

(1984). The inner circle: Large corporations and the rise of business political activity in the U.S. and U.K. New York: Oxford University Press.

61.

Van Waarden

. (1992). Emergence and development of business interest associations: An example from the Netherlands. Organization Studies, 13, 521-562.

62.

Williamson

(1968). Wage rates as a barrier to entry: The Pennington case in perspective. Quarterly Journal of Economics, 85, 85-116.