Week-15

Academy of Management Review. 1981. Vol. 6. No. 1, 145-154

A Contingency Approach to Labor Relations Strategies^

JON PROOSLIN GOODMAN University of Houston

WILLIAM R. SANDBERG University of Georgia

Our purpose in this paper is to identify the link between various behaviors involved in labor relations and organizational strategies. To this end, we have developed a typology and model of union/management behavior. The typology identifies companies according to the extent of unionization of the company and industry and according to union characteristics, and the model predicts labor relations strategy according to the match between company resources and union aggressiveness.

It is a commonly held belief among labor relations experts that "good" labor relations are indispensable to organizational success. There are many con- flicting suggestions offered to the firm seeking a labor relations strategy that will ensure "good" labor relations and thereby encourage success. In light of the scarcity of comprehensive theoretical models, practical descriptions or theory-based prescriptions, we intend to lay the foundation for a theoretical analysis of the interaction between labor relations strategies and organizational effectiveness. From this foundation, a contingency theory could be developed to arrive at a situationally based prescription for an organization's labor relations strategies.

Several steps must be taken to build a workable contingency theory. We have modified three steps that Hofer [1974] identified as being particularly im- portant; (1) the identification of measurable labor relations variables that significantly influence organizational effectiveness; (2) the development of a system for precisely describing an organization's labor relations strategy; and (3) the statement of hypothesized relationships between these variables and strategies. (A fourth step, the development of indices for measuring organizational effectiveness,

'We wish to express our gratitude for the inspiration and help of the late William F. Glueck in the developmental stages of this work.

© 1981 by ihe Academy of Management 0363-7425

is also necessary but is beyond the scope of this article.)

A firm's labor relations strategy is only one of several functional strategies that support its business and corporate strategies. By labor relations strategy, we mean the set of policies and techniques used by management to deal with its labor force and the goals and objectives they are designed to reach. Although many labor relations variables have been identified and measured, relatively little research has been done on strategies that contribute *o the measured behaviors.

A labor relations strategy can be categorized; accommodation, confrontation, co-optation, and collusion suggest themselves as appropriate generic labels. Components of these generic strategies might include the range of acceptable benefit packages, wage levels, and the perks and "extras" management uses as a basis for action or negotiations.

The following conceptual framework is directed at investigating the assumptions underlying such strategic types in order to permit an analysis of a company's position as if it had no labor relations program, no set strategy. Far too often an organiza- tion's approach to labor relations is the result of ac- cretion, and the strategies and behaviors that result are as much a function of tradition as of investiga- tion and analysis. Mainly because of the paucity of research into the effects of these strategies and policies, it is necessary to view what an organi-

145

zation does in the area of labor relations as if each strategem had to be justified, rather than automati- cally carried from one contract term to the next. In other words, just as zero-base budgeting requires re- evaluation and justification of expenditures for each new budget period, so should managerial planning. Zero-base managerial planning, to coin a phrase, is essential to the study of a company's labor relations program.

This is not to say that the labor relations strategy of a given firm should be justified solely by econo- mic criteria. Many issues are embedded in the tradi- tions built up by long association between an organization and its unions. Even so, this does not and should not prevent a company from evaluating all the components of its labor relations program in light of benefits that accrue or are presumed to ac- crue, whether those benefits are purely quantitative or both quantitative and qualitative. There are in- stances in which traditions are important, not only for their importance to continuing relationships, but also because jettisoning those traditions would have more dysfunctional effects than retaining them.

There are extensive bodies of literature on both labor relations and organizational effectiveness. However, there is little literature linking the two areas to define and measure organizational effective- ness within a labor relations context. And there is no consensus on defining and measuring organizational effectiveness in broader contexts. These complex issues are beyond the scope of this article. We will limit our focus to labor relations strategies, their conditional outcomes, and the sometimes uncertain relationships between these outcomes and measures of organizational effectiveness.

Defining the Issue

In order to measure the real costs and benefits of labor relations strategies and outcomes, one must define and bound the relevant concepts. Moreover, the issues must be addressed within a carefully delimited, readily researched context if the answers are to be useful. The United States manufacturing firm is the context we have adopted for this inquiry.

Although Western capitalist nations possess a generalizable type of labor relations system [Cox, 1977], there remain significant differences in the

nature of labor relations that are attributable to social distance, itself a product of a nation's educa- tional system [Brossard & Maurice, 1974]. Even within a single nation there are limits that must be imposed if the work and model are to be manage- able. In this article we focus on the manufacturing sector for several reasons besides simplification. Unionization is most intensive and mature in this sector, so data are more readily available and reflect greater stability than in other sectors. Also, the manufacturing sector already has an array of gener- ally accepted "hard" criteria of effectiveness. This is especially true at lower organizational levels, where dollars are usually a satisfactory measure of effec- tiveness and where questions of ultimate purpose and criterion do not intrude.

Choosing the firm as the unit of analysis is neces- sary if one is to develop a contingency theory prescribing organizational action for the firm. This choice departs from a major stream of industrial relations research that has concentrated on the func- tioning of a national labor relations system established by statute, regulation, and judicial precedent, rather than on the actions and outcomes of individual firms [Dunlop, 1958; Dunlop & Chamberlain, 1967; Kerr, Harbison, Myers, & Dunlop, 1960; Selekman, 1947]. One cannot ignore the importance of the national system, but changes in its effects are gradual unless there is a major legis- lative revision such as occurred in 1947 and 1959. To guard against discontinuities in labor law that would not be adequately captured in the model pro- posed below, research should be focused on periods between such changes.

There is an extensive body of literature addressing labor relations at the firm level. Most of it falls into two categories: exhortatory literature, calling for cooperative, "good" labor relations as a sure path to harmony; and general advice, often derived from personal experience and managerial tradition, on how any or all companies should deal with labor. Neither type approaches the issues of cost and benefit that are raised here, nor do they attempt to establish a clear, quantifiable, causal relationship between labor relations strategies and their effec- tiveness.

Typical of the exhortatory literature is this com- ment by Phillips: "Companies that get aggressive, harassing unions deserve them. . . . Any union will

146

pursue a strategy that is sensible in the environment first set by management" [1977, p. 69]. Phillips decried the traditional adversary relationship be- tween labor and management and urged that management take the lead in breaking down in- grained behefs that such a role is appropriate. His prescriptions are tactical rather than strategic (e.g., allow union meetings on the company premises to encourage participation by less motivated, and therefore less militant, union members; confer with union officials to learn the union's objectives so that negotiations can be more mutually beneficial) and they appear to accept as an article of faith a serial link connecting good interpersonal relations, good labor relations, and desirable organizational out- comes. A similar assumption underlies the prescrip- tions of Selekman [1947], although his human- relations-based approach was designed to bring "social good," as a step beyond desirable organiza- tional outcomes.

A large segment of the prescriptive category is exemplified by Wilson's forthrightly titled "Thoughts on Union Avoidance" [1977]. His recom- mendations are based on his lengthy experience with General Electric and are strongly reminiscent of Boulwarism. But like much work of this genre, Wilson's advice does not attempt to differentiate among types of firms or among the various situa- tions a given firm may face. One example of thoroughness and detailed prescription is found in Lawson [1968]. Although designed for the practi- tioner, this work offers unusually thorough analyses of union organizing strategy and management's potential vulnerability. Lawson's prescriptions con- stitute a comprehensive program for defeating a union in a recognition election, but do not discriminate among situational variables or attempt to measure the consequences of unionization (or of his recommended campaign) for organizational effectiveness.

We have found few clear attempts to use a con- tingency approach in labor relations, and none that provided for labor relations strategy as an indepen- dent variable or determinant. Gerhart [1976] based his model on union and management bargaining- power ratios drawn from Chamberlain and Kuhn [1965], which express each party's power as the ratio of the cost of disagreeing to the cost of agreeing. Gerhart used eighteen independent variables to predict bargaining outcomes in local government

labor negotiations. Only half of the variance in the dependent variable (contract analysis from labor's standpoint) was explained by the eighteen vari- ables. Gerhart believed that the large error term was attributable to the static, cross-sectional nature of his data and to the model's omission of labor market conditions, transient political conditions, the historical development of the relationship between the parties, and their bargaining strategies and abilities.

Koch and Fox [1978] sought only to develop test- able propositions on the likely degree of pressure for worker participation in corporate decision making. Their work resembles this one in its focus on con- structing a conceptual model and preparing for em- pirical investigation. They identified five organiza- tional forces and seven dimensions of the industrial relations setting that mediate the pressure for in- creased participation. Koch and Fox ventured far- ther than most along the way to a contingency theory, but proposed no criteria or measurement for the identified variables.

The criteria themselves would best be developed from existing work in the area of organizational effectiveness. Although there is a significant body of literature on the subject, many researchers agree that there is no single criterion of effectiveness and that, as currently understood, methods of assessing effectiveness will vary from organization to organi- zation [Cunningham, 1977; Kirchoff, 1977; Yucht- man & Seashore, 1967].

The scope of the evaluation criteria that have been suggested to measure organizational effectiveness is considerable, ranging from profitability, growth, and survival [Child, 1974, 1975; Negandhi & Riemann, 1973] to highly complex, integrated models that combine economic factors such as profit and productivity with behavioral measures such as employee satisfaction, societal value, cohesion, and interpersonal relations [Friedlander & Pickle, 1968; Georgopoulous & Tan- nenbaum, 1957; Price, 1968; Webb, 1974]. Macy and Mirvis [1976] have made great strides in quanti- fying the economic impacts of such behaviors as strikes and work stoppages, grievances and acci- dents, absenteeism and tardiness, turnover, and productivity as a step toward establishing equiv- alent measures.

In the context of labor relations, it is beneficial to view the criteria used to measure organizational

147

effectiveness as a subset of a larger, more com- prehensive model that would measure how well a particular organization integrates its actions with the constraints and opportunities posed by its task environment. Thus, the measure of a company's ability to deal effectively with its labor force would be based on situational prescriptions derived from its own internal and external profile.

Our proposed model is an attempt to define the labor relations strategy of management and the col- lective behaviors and resources of the company's workforce, so that a prescriptive theory of labor relations strategies for management can be developed.

The Model and Typology

When examining a concept as amorphous as labor relations, one is often governed by convenience in selecting a starting point. We have chosen to begin with a matching of company resources and union aggressiveness.

The matrix in Figure 1 is labelled "Company Resources" and "Union Aggressiveness" rather than the converse (company aggressiveness, union resources) for several reasons. First, "Company Resources" is a major independent variable because of the nature of for-profit enterprise. Few firms are founded on ideological or political grounds; rather, resources are generally their central concern. The extent of company aggressiveness thus is limited by the company's financial condition. Few companies would take an aggressive stand that promised finan- cial ruin. (It seems likely that the aggressive posi- tions of J.P. Stevens and Kohler are based on management's belief that some long-term organiza- tional gain will result from their policies.)

A second reason for our choice of labels is that company aggressiveness is a product of many other factors besides the company's financial condition. Within the limits imposed by available resources, management may give rein to whatever innate aggressiveness or anti-union ideology it harbors. These factors will influence the choice among avail- able options, and therefore can be conceived as operating on rather than in our model. Our inten- tion is to help management understand and predict the outcomes of various labor relations strategies.

not to predict their choices. Our use of "Union Aggressiveness" results from

parallel reasoning. Although unions operate under financial constraints, these are not the dominant fac- tors in their strategies. Resource accumulation is not the primary objective and unions generally satisfice in the pursuit of financial ends. The economic burden of a strike may be borne by union members as well as by the union's strike fund. Members' will- ingness to sacrifice family savings will depend on their loyalty to their peers and to the union as an in- stitution. Thus "union resources" only partially ex- plain union aggressiveness, and are a component of larger union strategies that might influence a union's actions. Many financially secure unions are relative- ly passive, whereas impoverished unions are often militant.

Company Resources'"

High

Medium

Low

Union Aggressiveness'

High Medium Low

1 strike likely

4

manage- 7 ment yields

2

5

•8

labor yields

6

accommo- dation

Figure 1 Matching Company Resources

And Union Aggressivenes

^Union aggressiveness: cell assignment of subject union would depend on grievance frequency, settlement ratio (proportion oF contract negotiations marked by strikes), and work stoppage and work slowdown frequencies. Tfie actual rates determining fiigh, medium, and low aggressiveness will be fixed after an analysis of union behavior, wfiich will provide a relevant context for evaluation. (Tfiese measures are for the union s dealings with all tfie firms in tfie industry except Ifie one being classified.)

'Company resources: cell assignment of subject firm would depend on ex- istence of mutual-aid pacts within industry: ability of company to with- stand shutdown in terms of short-term debt, fixed costs, inventory depth, vulnerability of market share, vulnerability to substitute goods; and other industry-specific costs of discontinuity.

r n Identifies firms in situations that have relatively unpredictable out- comes.

148

Company Resources '

A firm's bargaining strength in relation to its union (s) depends on its financial and strategic posi- tions. Financial conditions that are important to a firm's labor relations strategy include unavoidable cash obligations versus cash resources, and the general financial strength of the firm. A company's ability to cover fixed costs for the period of a strike or slowdown will have a major influence on its labor relations strategy.

Strategic considerations, such as the industry's stage in the product life cycle and the firm's com- petitive position within the industry, are also signi- ficant. A firm's ability to tolerate discontinuity is reduced by the ready availability of competitors' or substitute goods, especially if market share is im- portant and customer loyalties are ephemeral.

Management can mitigate both financial and strategic weaknesses by the identification of a sus- taining core. Ideally, this core would buffer a firm's strategic position by using managerial personnel to continue those operations which are central to the company's market position or cover a significant proportion of its unavoidable obligations at an ac- ceptable cost in diverted management attention. A temporary change in the scope or method of opera- tions, using supervisory or management personnel, might increase a company's ability to cope with labor problems. In some cases, unions will perceive such operations as legitimate, and the company will not face lasting union animosity.

Thus, assignment of a firm to a "company re- source" cell will be made after analyzing the firm's financial constraints, its strategic position, and the relationship between the two. A strong financial condition will enable a company to have greater strategic latitude, just as a strong strategic position can increase financial flexibility.

Union Aggressiveness

One widely discussed and measured dimension of aggressiveness is militancy. Its contribution to a union's willingness to strike is so well accepted that strikes are often used to measure militancy [Fox & Wince, 1976; Ostrander, 1970; Shirom, 1977]. Atti- tudinal measures of militancy are most commonly

derived from responses of union members to ques- tionnaires or interviews. This method would be almost impossible to use in the context most signifi- cant to corporations, however. After all, manage- ment would rarely be able to sample union members and might be legally barred from doing so in many instances. Therefore, despite its seeming value, this type of measurement is of little use for our purpose.

Instead, the proposed model uses several behav- ioral variables that do not require administration of a research instrument. There is support in the literature for using work stoppage frequency [Shirom, 1977], slowdowns [Kovner & Lahne, 1953; Seidman, London, Karsh, & Tagliacozzo, 1958], arbitrated grievances [Hoellering, 1975], and wildcat strikes [Zack, 1972].

It should be noted that union aggressiveness has been defined here to exclude all data concerning the union's dealings with the subject firm. The behavioral variables (see Figure 1) are to be measured for the union's dealings with the entire in- dustry (if practicable) except for the firm being classified. This precaution prevents a built-in bias in defining and measuring effectiveness, because many of the behaviors used as indicators of union aggres- siveness will, at the firm level, be criteria of effec- tiveness. To include data on the union's relations with the subject firm would predispose this model to find an inverse relationship between union aggres- siveness and organizational effectiveness. Instead, the measures of aggressiveness will serve as an in- dustry standard against which the subject firm's unions can be compared.

The Company/Union Matrix

Company resources and union aggressiveness each have been divided into three states (high, medium, and low), resulting in the nine possible combinations depicted in matrix form in Figure 1. The four corner cells of the matrix represent situa- tions that produce relatively predictable outcomes: either one party is clearly dominant and the other yields, as in Cells 3 and 7, or the parties are at a standoff, as in Cells 1 and 9. In the latter cases a strike is likely if extreme union aggressiveness confronts strong corporate resources, while accommodation is likely if union passivity coincides with low cor-

149 I

porate resources. The five remaining cells (shaded area) represent situations in which there is less cer- tainty.

The four corner cells are less germane to this in- quiry than the five other cells. They are important, but present less strategic challenge to management. Although an organization may be profoundly affec- ted by what transpires within these cells, manage- ment's decision making has been reduced from the strategic to the tactical level by the virtual removal of uncertainty. Thus, the firm in Cell 1 may face crucial decisions during a long, arduous strike, but these decisions generally will involve tactical im- plementation of a strike plan rather than the strategic decision to endure the strike. The labor relations literature is replete with examples, advice, and rules for conducting strikes without running afoul of NLRB regulations, and companies are more apt to consult their lawyers than their strategic plan- ners during a strike.

The five cells in which uncertainty prevails are our primary concern. Managers encountering such situations of uncertainty face several key strategic decisions. They must decide whether to accede to demands or endure a strike. Then they are likely to face additional choices on whether to relent, com- promise, or stand firm on specific bargaining points. Intelligent, informed action here depends in part on management's ability to identify the costs and benefits associated with each condition of employ- ment, managerial prerogative, fringe benefit, or other item on the negotiating agenda. If manage- ment's knowledge of the costs of their choices is defi- cient, strategic decisions are more likely to be faulty: too many or too few strikes will be taken; too much or too little will be conceded to avert strikes.

The knowledge gathered for making strategic decisions by companies assigned to the uncertainty cells would be useful to companies in the certainty cells, who face essentially tactical decisions, as well as to companies in the uncertainty cells making tac- tical decisions in pursuing their chosen strategic plans. Yet this dual utility does not contradict the distinction made above between strategic and tac- tical decision making. Knowing the stakes enables managers to make better strategic decisions even without choosing precisely the points on which they later may compromise. Information is relevant only to the extent that it is useful in strategy formulation

and selection. There is no intrinsic value or interest in the information.

Labor Relations Contexts

In an attempt to reduce the uncertainty inherent in Cells 2, 4, 5, 6, and 8, we have devised a typology of firms and industries. Industries are classified as non- union, partially unionized, or unionized. In addi- tion, three major union characteristics have been defined: level of organizing activity, cohesion, and resources. These variables are the building blocks of the contingency theory. The symbols, definitions, and measures for these variables can be found in Table 1. For example, a company could be "III-2," dealing with an "ACr" union. These types, inserted into the appropriate cells of the matrix in Figure 1, would be the beginning of the development of a situationally based prescription for management ac- tion.

Most of the definitions presented in Table 1 are straightforward and the criteria associated with them are arbitrary. The intent at this stage is to develop and refine the measures and ranges that have been chosen. The values designated as cutoff levels are not necessarily those which the model would eventually use, but are suggested as reason- able starting points.

Most of the terms used in Table 1 are clear, but two warrant elaboration and perhaps justification. Union cohesion refers to supportiveness among locals and compliance by locals with the national union's positions. This characteristic is included as a bridge between the national union and the relevant local(s). Insofar as unions themselves have strategies that are analogous to those of firms, we expect that national strategies are executed unevenly in different regions, industries, or other contexts. The degree of cohesion among locals and between locals and the national union may have an important effect on a local's actions when national strategy changes. For example, a local previously identified as passive and with a tradition of high cohesion may become more militant as a result of the increased militance of a new national strategy.

In one respect the treatment of cohesion is similar to this model's treatment of union aggressiveness: the bulk of the literature on cohesion relies on

150

Table 1 Typology of Industries, Companies, and Unions

Type of companies within types of industries

Non-union company Partially unionized company Union company

Non-union industry Partially unionized industry Union industry

= I = II = III

= 1 = 2 = 3

Union Characteristics

Organizing activity High Low

Cohesiveness High Low

Resources High Low

A a

C c

R r

Definitions

Non-union industry: Less than 15% of sales gener- ated by union companies; or less than 15% of total workforce organized.

Partially unionized industry: 15-85% unioniza- tion, either by sales or workforce.

I Unionized industry: More than 85% unionization, either by sales or workforce.

The seeming indecision between sales and work- force reflects industry conditions. Some are labor intensive, whereas others generate large revenues from highly automated plants. The important fac- tor here is consistency; the same criterion must be applied throughout an industry.

Union organizing activity: At least n attempts to organize a company or unit in the industry within the last k years implies high union organizing ac- tivity (because union activity and its effect vary both by union and by industry, this definition does not specify values for n and k).

Union cohesion: Percentage of settlement rejec- tions, agreement of local with national (evidenced by number of wildcat strikes), and willingness of locals to support strikes of other locals. Once again, what is high cohesion in one union within an industry does not necessarily mean high cohesion in another union or industry.

Union resources: Resources are considered high if the strike fund and available state unemployment compensation together provide two-thirds of the pay that was earned while working, for twelve weeks.

151

attitudinal measures that are usually unavailable to management. Yet a workable contingency theory depends on the decision makers' ability to identify and measure situational variables. Therefore we rely once more on overt behavior rather than attitudes. The measures suggested in Table 1 are readily avail- able and data can be interpreted in light of historical standards for the industry or union.

The term union resources also requires elabora- tion. Like company resources, union resources in- fluence the ability to endure a strike, although their role is not as decisive. The union's strike fund serves the same purpose as strike insurance or mutual aid pacts. Unemployment compensation is available to strikers in some states and is assumed to reduce their opportunity costs. As elsewhere in this model, we have begun with an arbitrary standard in Table 1 — which we would modify without hesitation in order to enhance its discriminatory effectiveness.

Once the company/industry/union label has been derived in Table 1, it is placed in the appropriate company/union cell in Figure 1. This creates the classification foundation necessary for a situational- ly based prescription for management action. There are 72 possible permutations of contextual variables (i.e., labels) within each of the nine cells in the com- pany/union matrix. As research progresses it is like- ly that some permutations can be combined to reduce the model's complexity; at this early stage, however, one does not know which characteristics will prove most significant in which cells. The most important caveat at this level of theory building is that research should be done on a cell-by-cell basis, thus reducing the considerable number of variables that would have to be dealt with if one were to in- vestigate using the matrix in toto.

External Variables

The above components of the model have been described in ways that might lead researchers to assume that we have ignored the vital external en- vironment of the organization or union. At this stage of theory development, however, external variables such as the state of the economy, market factors (shortage or glut), transient governmental antitrust policy, or availability of resources must be kept "on the back burner."

After some attempt has been made to opera- tionalize different cells and investigate the behaviors that accompany the situations they describe, the ex- ternal variables could be examined as possible mediators of the prescriptive results. First the con- tingency model should be developed, the cells classified and described, and then when the model is more advanced, the environmental factors could be examined to determine the effect they might have on situations that are internal to the firm or the union with which the firm must interact.

Considerations and Caveats For Future Research

The theoretical basis of this contingency approach is the assumption that "symptoms" associated with the domain of labor relations are, in fact, measures of organizational effectiveness. It must be remem- bered, however, that they measure only certain aspects of total organizational effectiveness, and cannot be taken as indicators of overall perfor- mance. There are many criteria and they can be ex- pected to vary over time, among industries, and according to specific circumstances. In other words, the same criteria may not be relevant in the steel in- dustry in both 1948 and 1978, or in both the steel and meatpacking industries in 1978.

The many measures of organizational effective- ness, and the extensive literature attesting to the in- appropriateness of using a single criterion such as earnings, indicate the importance of considering a range of qualitative and quantitative measures. Price [1968], Kirchoff [1977], Cunningham [1977], and Smith [1976] are among many who have pointed to the need for multiple measures of organi- zational effectiveness, which must be determined ac- cording to their relevance to the particular organiza- tion's context, goals, or problems.

According to Dubin, an applied scientist attempts to construct models to explain symptoms, while a practitioner "usually employs a pre-existing model to indicate how the symptoms should be treated" [1976, p. 20]. We have sought to describe conditions that are at once a list of symptoms and (when mea- sured by degree or by frequency of occurrence) in- dications of organizational effectiveness. The hard criteria are costs that are measurable in dollars

152

(Imberman, 1979; Macy & Mirvis, 1976] or in units of output, and that are usually available from com- pany records. They might include the frequencies of strikes, other job actions (slowdowns, stoppages, or sickouts), sabotage, grievances, and grievances taken to arbitration. The soft criteria, while measurable, are more amenable to analysis in qualitative and subjective terms. Frequently the ambiguity of standards or an insufficient under- standing of causality forces the evaluator to aban- don the hard standard of efficiency in favor of relative or instrumental standards [Thompson, 1967]. Thus, the soft criteria are likely to be exam- ined through a combination of company records, direct observation, expert opinion, and other alter- natives to hard criteria. Some possible soft criteria are the resolution of health and safety issues, the retention of management prerogatives, and the ef- fect of outcomes on the company's ability to execute its business strategy and on its flexibility in modify- ing that strategy.

At this early stage, progress toward a contingency theory of labor relations would be advanced even by

research that makes no pretense of approximating ideal experimental conditions. To limit the problems of complexity and ambiguity, we would select firms within a single industry and, to the extent possible, within a single cell of the matrix in Figure 1. This should reduce the effects of variables not captured in this model. For the volume and richness of data re- quired on each organization studied, there appears to be no alternative to a longitudinal case approach. The longitudinal nature of the data would permit stronger inferences of causality than would cross- sectional data.

Managements use many labor relations tactics. Despite the universal prescriptions offered by some writers, there is no agreement in the literature on whether symptoms share a common cause or are remedied by a common treatment. Nor has any tac- tic emerged as the preferred remedy for a particular symptom. Here, in the relatively uncharted realm of conditional causality, a contingency theory could contribute to understanding, and ultimately to mak- ing effective action possible according to prescrip- tion.

REFERENCES

Brossard.M.; & Maurice, M. Existe-t-il un modele universel des structures d'organisation? Sociologie du Travail. 1974, 16, 402-426.

Chamberlain, N . ; & Kuhn, J. Collective bargaining (2nd ed.). New York: McGraw-Hill, 1965, pp. 170-171; cited in Gerhart, 1976.

Child, J, Managerial and organizational factors associated with company performance. Joumal of Management Studies. Part I: 1974, 11, 175-189; Part II: 1975, 12, 12-27.

Cox, R. W. Pour une fetude prospective des relations de produc- tion. Sociologie du Travail, 1977, 19, 113-137.

Cunningham, J. Approaches to the evaluation of organizational effectiveness. Academy of Management Review, 1977, 2, 463-474.

Dubin, R. Theory building in applied areas. In Marvin Dun- nette (Ed.), Handbook of industrial and organizational psychology. Chicago: Rand-McNally, 1976, pp. 17-39.

Dunlop, J. Industrial relations system. New York: Holt, 1958.

Dunlop, J; & Chamberlain, N.W. (Eds.). Frontiers of collective

bargaining. New York: Harper & Row, 1967.

Farmer, G. The use of injunctions in labor disputes: A manage- ment view. In Richard Adelman (Ed.), Proceedings of New York University 30th Annual Conference on Labor. New York: Matthew Bender, 1977, pp. 191-200.

Fox, W.S.; & Wince, M.H. The structure and determinants of occupational militancy among teachers. Industrial & Labor Relations Review, 1976, 30, 47-58.

Friedlander, F.; & Pickle, H. Components of effectiveness in small organizations. Administrative Science Quarterly, 1968, 13, 289-304.

Georgopolous, Basil S.; & Tannenbaum, Arnold S. A study of organizational effectiveness. American Sociological Review, 1957, 22, 534-540.

Gerhart, Paul F. Determinants of bargaining outcomes in local government labor negotiations. Industrial & Labor Relations Review, 1976, 29, 331-351.

Getman, J.G.; Goldberg, S.G.; & Herman, J.B. Union represen- tation election: Law and reality. New York: Russell Sage Foun- dation, 1976.

153

Hendricks, W. Conglomerate mergers and collective bargain- ing. Industrial Relations, 1976, 15, 57-85.

Hirsch, J.S. Strike insurance and collective bargaining. In- dustrial & Labor Relations Review, 1969, 22, 399-415.

Hoellering, M. Expedited arbitration. In Richard Adelman (Ed.), Proceedings of New York University 28th Annual Con- ference on Labor. New York: Matthew Bender, 1975, pp. 319-335.

Hofer, C. Research on a contingency theory of strategic behavior: Issues and methods. Paper presented at the Midwest Meeting of the Academy of Management, April 1974.

Imberman, W. Strikes cost you more than you think. Har- vard Business Review, 1979, 57, 133-138.

Kerr, C ; Harbison, F.; Myers, C ; & Dunlop, J. Industrialism and industrial man. Cambridge: Harvard University Press, 1960.

Kirchoff, Bruce A. and policy research. 347-355.

Organization effectiveness measurement Academy of Management Review, 1977,2,

Koch, J.L.; & Fox, C.L. The industrial relations setting, organizational forces, and the form and content of worker par- ticipation. Academy of Management Review, 1978, 3, 572-583.

Kovner, J.; & Lahne, H. Shop society and the union. In- dustrial & Labor Relations Review, 1953, 7, 3-14.

Lawson, J.W., II. How to meet the challenge of the union organizer. Chicago: Dartnell Corporation, 1968.

Mabry, B. The economics of fringe benefits. tions, 1973, 12, 95-106.

Industrial Rela-

Macy, B.A.; & Mirvis, P.H. A methodology for assessment of quality of worklife and organizational effectiveness in behavioral economic terms. Administrative Science Quarterly, 1976, 21, 212-226.

Morel, C. Les strategies de nfegociation dans l'entre- prise. Sociologie du Travail, 1977, 19, 362-382.

Neghandi, A.R.; & Reimann, B.C. Task environment, decen- tralization, and organizational effectiveness. Human Relations, 1973, 26, 203-214.

Northrup, H.R. Comment. Review, 1977, 30, 364-372.

Industrial & Labor Relations

Odewahn, C.A.; & Krislov, J. Contract rejections: Testing the explanatory hypothesis. Industrial Relations, 1973, 12, 289-296.

Ostrander, K.H. Measurement of teacher attitudes toward sanctioned activities, foumal of Education Research, 1970, 63, 219-221.

Peterson, R.B.; & Tracy, L. Testing a behavioral theory model of labor negotiation. Industrial Relations, 1977, 16, 35-50.

Phillips, M. Developing healthy management attitudes for deal- ing with a union. Personnel, 1977, 54, 68-71.

Price, J.L. Organizational effectiveness: An inventory of pro- positions. Homewood, 111.: Irwin, 1968.

Seidman, J.; London, J.; Karsh, B.; Tagliacozzo, D.L. The worker views his union. Chicago: University of Chicago Press, 1958.

Selekman, B.M. Labor relations and human relations. York: McGraw-Hill, 1947.

New

Shirom, A. Union militancy: Structural and personal deter- minants. Industrial Relations, 1977, 16, 152-162.

Smith, P.C. Behaviors, results, and organizational effec- tiveness: The problem of criteria. In Marvin Dunnette (Ed.), Handbook. of industrial and organizational psychology. Chicago: Rand-McNally, 1976, pp. 745-770.

Stein, B. Economic policy and collective bargaining. In Richard Adelman (Ed.), Proceedings of New York University 29th Annual Conference on Labor. New York: Matthew Bender, 1976, pp. 147-150.

Thompson, J.D. Organizations in action. McGraw-Hill, 1967.

New York:

Unterberger, S.H.; & Koziara, E.G. Airline strike insurance: A study in escalation. Industrial & Labor Relations Review, 1976, 30, 26-45.

Webb, Ronald J. Organizational effectiveness and the volun- tary organization. Academy of Management Joumal, 1974,17, 663-677.

Wilson, J. Thoughts on union avoidance. Personnel Ad- ministrator, 1977, 22, 14-18.

Yuchtman, Ephraim; & Seashore, Stanley E. A system resource approach to organizational effectiveness. American Sociological Review, 1967, 32. 891-903.

Zack, A.M. Impasses, strikes, and resolutions. In Sam Zegoria (Ed.), Public workers and unions. Englewood Cliffs, N.J.: Prentice-Hall, 1972.

Jon Prooslin Goodman is an Assistant Professor of Management at the University of Houston.

William R. Sandberg is a doctoral candidate in business policy at the University of Georgia.

Received 4/9/79

154