Economic Behavior Of Trade Unions Research Paper

View sample Economic Behavior Of Trade Unions Research Paper. Browse other  research paper examples and check the list of research paper topics for more inspiration. If you need a religion research paper written according to all the academic standards, you can always turn to our experienced writers for help. This is how your paper can get an A! Feel free to contact our custom writing service for professional assistance. We offer high-quality assignments for reasonable rates.

1. Trade Unions

A trade union is an organized association of workers formed for the protection and promotion of their common interests. The standard view of trade unions is that they are monopoly organizations that improve the welfare of members, principally by raising wages above the competitive level. This research paper investigates the conditions under which a union can increase wages, and explores ways of modeling the competing preferences of unions and management.

2. Conditions Under Which A Union Can Improve Workers’ Welfare

For a trade union to be able to increase wage rates above the competitive level, there must be some surplus that can be shared between the firm and the union, and the union must have some bargaining power to induce the firm to share this surplus. How can a union achieve such power? It may be able to organize all workers in an industry, thereby acting as a monopolist over the supply of labor. The earliest successful unions were craft unions, which became established because of a combination of high demand for skilled labor during industrialization, and the control of trained labor by skilled workers through the apprenticeship system. The emergence of general or industrial unions followed a different path. During industrialization, the huge surplus of unskilled displaced agricultural labor made manipulation of labor supply impossible. A depression, coupled with a readily available pool of substitute workers, could destroy a union. To ensure survival, a general union needed political support or very high levels of membership. Obtaining the latter was difficult in the early stages of unionization, when there were no immediately obvious wages benefits because the union had not yet obtained any power.

Union bargaining power derives from the ability of the union to inflict damage on the firm by withdrawing labor. This works when there is no substitute pool of nonunion workers, for example, if there are high labor turnover costs or if the union controls labor supply through entry restrictions or through having high levels of membership relative to the size of the sector. The view that workers have to battle to acquire sufficient power to induce the firm to concede any surplus accords with the Marxist analysis of the capitalist means of production (see, for example, Bowles 1985). Union membership density may play a part in determining the power of the union to extract from the firm a share of any rents. As membership density increases, there is a shrinking pool of workers available to the firm should it choose not to employ union labor. Thus one would expect to find a minimum critical level of membership below which firms would refuse to recognise and bargain with the union (Osborne 1984). Where membership density is crucial to the union’s monopoly power as, for example, with the emergence of the general unions, the union must consider how to attract workers at the margin. The basic problem for economists in explaining why workers join a union relates to the free-rider problem, arising because of the collective nature of many of the goods provided by the union (Olson 1965, Booth 1985). Why join the union when union dues are costly, and when the benefits are not immediately available or apply to all workers regardless of their union status? Of course established unions may also provide incentive private goods that encourage workers to unionize. Indeed, a union can act as the workers’ agent, since it can reduce the costs of negotiation and processing of grievances, a role which may also be useful to employers (Faith and Reid 1987).

Even if a union controls labor supply, it will not necessarily be able to negotiate a large wage increase relative to the competitive level. The magnitude of the union wage depends crucially on the elasticity of labor demand in that sector. There can be little doubt that, if unions emerge in competitive markets, high union wages introduce allocative inefficiencies into the economy through the distortion of factor prices. Without unions, allocative efficiency is associated with the equalization of the marginal products of identical factor inputs across sectors. With higher wages in the union sector, union firms employ fewer workers. Displaced union workers crowd into the nonunion sector, lowering wages there. As a result, too few workers are employed in the union sector where output falls, while too many workers are employed, and too many goods produced, in the nonunion sector. There is a deadweight efficiency loss, because the value of marginal products in the two sectors are not identical. In addition, there are distributional issues to consider, and longer run effects due to substitution of capital for labor. Furthermore, in unionized sectors there may be under-investment in capital through the hold-up problem (see, for example, Grout 1984).

This discussion of allocative inefficiency assumes that the union has emerged in an economy characterized by competitive product and labor markets. But there is considerable evidence that, in modern industrialized countries, many product markets are characterized by imperfect competition. Moreover, nonunion firms may also face an incumbent workforce with a degree of bargaining power (Lindbeck and Snower 1988). Thus, even in the absence of unionization, management and workers may be in a situation of bilateral monopoly. An important question (little addressed in the literature) arises as to whether or not the replacement of individual bargaining by collective bargaining generates additional inefficiencies and misallocation of resources in situations where markets were previously not functioning in accordance with the textbook model of perfect competition. The majority of formal trade union models in the literature assume a perfectly competitive product market to allow the models to focus on wage and employment determination in the simplest environment (Dowrick 1989 inter alia is an exception). Nonetheless, it is an empirical regularity that imperfections in product and labor markets are correlated (Stewart 1990).

Even within the competitive model, there are arguments suggesting that, in the presence of imperfect information and uncertainty, unions may enhance efficiency. To the extent that unions reduce labor turnover and negotiating costs, they may increase the available surplus to be shared between parties (Freeman and Medoff 1984). Of course, there may be interdependence between the monopoly and efficiency roles of trade unions: unless the union has some bargaining power, it may be unable to increase efficiency.

3. Formal Modeling Of Union Behavior

To investigate union behavior, the economist needs to model the objectives of the relevant actors involved in wage setting, as well as the means of resolving their potentially conflicting objectives. Most union models consider the objectives of two agents—the union and management. Typically, the government’s behavior is assumed exogenous. Because the objectives of union and management may be conflicting, models need to consider how conflicts might be resolved within the constraints provided by the institutional environment. The range of issues covered by collective bargaining in Europe is narrower than in the USA, which probably reflects a relative lack of US social welfare legislation. (In Europe, legislation typically covers issues such as health insurance, pensions, redundancy pay, and unfair dismissal that are often covered by collective agreements in the US.)

While bargaining may take place over a number of issues, for tractability researchers typically concatenate wages and other forms of compensation, such as fringe benefits, under the single heading of wages. Union preferences appear to include employment as well as wages, but it is rare to find instances of the union and the firm bargaining over both wages and employment. Of course, unions affect employment indirectly, not only through increasing wages but also through negotiation over staffing levels, recruitment, and redeployment.

While unions typically have a number of goals (from increasing remuneration to broad political aims concerning the labor movement as a whole), the orthodox union model assumes unions are concerned only with the unionized sector. But a union typically comprises heterogeneous individuals with varying and possibly conflicting preferences as to the union’s strategy. For example, there may be principal–agent problems arising because union leaders (agents) have objectives that may conflict with the membership (the principal). The leadership may be better informed than its membership, or perhaps interested in self-aggrandizement or in wider labor goals. Unions are usually large, with complex organizational structures concerned with representation of members’ interests and with administration. Because the internal organization of the union is typically hierarchical in structure, the question arises as to how well members’ preferences are transmitted through the various levels, each of which may have conflicting aims. Moreover, there is heterogeneity across unions in terms of their structure and constitution (ranging from the democratic through to the totalitarian) that may determine a union’s maximizing strategy.

While there is a small literature allowing for union objectives to be affected by membership heterogeneity or a conflict of interests between leaders and members (Farber 1978, Blair and Crawford 1984), the widely used models of union behavior side-step these problems by supposing that all workers in the union sector are identical and are union members. They also assume that the union cares only about the economic welfare of the union sector. For simplicity, union-negotiated benefits are assumed to be encapsulated in the wages variable. Because the union is also hypothesized to care about the employment of union workers, union objectives can be specified as an increasing function of wages w and employment n.

Economic Behavior Of Trade Unions Research Paper

In most recent union literature, the union is assumed to maximize either a utilitarian objective function, or an expected utility objective function. The utilitarian objective function assumes that the union maximizes the sum of the individuals’ utilities. The union comprises identical members treated identically by the union, and therefore in this context no distributional or normative judgments are being made by the utilitarian objective. However, if workers were assumed heterogeneous, the use of this as the union objective would introduce normative judgments (since if individuals were not identical, it would be necessary to consider why the utilitarian social welfare should be adopted instead of some alternatives such as the Rawlsian or Maxi–min criterion, or a more egalitarian approach). If the union’s wage-setting behavior results in some union members being unemployed, the utilitarian objective function is:

Economic Behavior Of Trade Unions Research Paper

where u(.) is a concave (indirect) utility function, w denotes wages, n denotes employment in the union sector, t is the total number of union workers in the sector (and membership by assumption), and b is the alternative sector wage or the unemployment benefit.

The second popular union objective function is the expected utility function. If membership is fixed, Eqn. (1) can be written equivalently in terms of expected utility. Since the union raises wages above the competitive level, each member faces some probability (1 – n/t) of being unemployed. If unemployed, a worker receives the alternative sector wage or the unemployment benefit. The expected utility of a representative union member is given by:

Economic Behavior Of Trade Unions Research Paper

where u`(w) = du/dw. The slope of the union indifference curve in (w, n) space is found by totally differentiating either Eqns. (2) or (3) with respect to w and n, holding (expected) utility fixed, yielding dw/dn = – [u(w) – u(b)]/nu`(w)

Both the utilitarian and the expected utility objective functions have the implication that identical workers receive different outcomes ex post. Employed members receive the union wage, while the unemployed workers receive b. However, if unemployed members effectively leave the union, the leadership will be concerned about future union viability, since there will be a larger pool of nonunion labor that might undercut the union wage, and union income from dues would fall. This is likely to affect union outcomes, and is an argument for inclusion of the level of membership in the union objective function as an endogenous variable. Moreover, b may not be exogenous—the union may affect remuneration for unemployed members through redistribution by benefits or redundancy pay, suitably adjusted for the disutility of work for those in employment. Indeed, historically, many unions paid unemployment benefits to out-of-work members, and some commentators argue that this was important for union growth and survival.

Now consider the behavior of the firm. The necessary conditions for a union to be able to appropriate any surplus in a competitive product market, without driving the firm out of existence, are: very high levels of unionization in a competitive industry, no nonunion foreign competition, entry barriers, and inelastic product demand. Without the first three conditions, domestic firms, or foreign firms, or new firms entering the industry could simply hire nonunion labor and produce at a lower cost than unionized competitors, thereby driving them out of business. Assuming the firm maximizes profits and has no power in the labor market (which means it will take wages as given), it has just one choice variable—employment. The optimization problem facing the firm is therefore:

Economic Behavior Of Trade Unions Research Paper

where π denotes profits, p denotes product price, and q(n) is the production function characterized by diminishing returns to labor. The firm’s wage bill is wn and capital is held fixed. The slope of the firm’s isoprofit curve in (w, n) space is found by totally differentiating Eqn. (3) with respect to wages and employment keeping profits fixed, to obtain dw/dn = [pq`(n) – w]n. The locus of the turning point of each isoprofit curve traces out the labor demand curve where pq`(n) = w, as illustrated in Fig. 1. Two isoprofit curves IP0 and IP1 are illustrated in Fig. 2, together with two union indifference curves IC0 and IC1.

Economic Behavior Of Trade Unions Research Paper

4. The Monopoly Union Model

In the monopoly union model, there is no bargaining between the two parties. Instead, the union acts as a monopolist in the supply of labor and imposes a wage rate on the firm. The firm, however, retains its managerial prerogative to determine the number of workers employed at the union wage. Once the union sets the wage, the firm reads off from the labor demand curve the number of workers to hire at that wage. Thus, a union maximizing the expected utility of its representative worker sets wages as follows:

Economic Behavior Of Trade Unions Research Paper

The first-order condition can be found directly by equating the slopes of the union indifference curve (giving the union’s marginal rate of substitution of employment for wages) with the slope of the labor demand curve, illustrated as point A in Fig. 2. The monopoly union seeks its highest level of utility, subject to the constraint of the firm’s labor demand curve, where w* and n* are the optimal wage and employment levels. It can be seen, from the first order condition in Eqn. (5), that wages will be set by the union such that the percentage increase in a member’s utility due to a percentage increase in wages is exactly equal to the wage elasticity of labor demand ε. Intuitively, this result arises because an increase in wages reduces union employment, and thus each worker faces a greater probability of being unemployed. On the other hand, any member lucky enough to be employed receives a higher wage. Thus, in equilibrium, the percentage marginal benefit is equal to the percentage marginal cost:

Economic Behavior Of Trade Unions Research Paper

This equation simply characterizes the equilibrium situation. To obtain a specific value for wages, it is necessary to choose explicit functional forms for the utility function and the labor demand function that can then be inserted into the generalized characterization and solved for the wage rate. For evaluation of empirical work attempting to test this model, see Pencavel (1991) and Booth (1995).

Using this model, it is straightforward to show how the wage set by the monopoly union alters due to a change in each of the explanatory variables. An isoelastic increase in labor demand has no effect on wages, but employment increases. An improvement in alternative opportunities increases the union wage, but employment is reduced as we move up the labor demand curve. An increase in membership has no effect on wages or the numbers employed. The empirically testable predictions of the monopoly union model are first that the alternative wage and unemployment benefit should affect wages directly. But employment should be determined by the wage rate, and unaffected by variables measuring alternative opportunities. Second, membership changes should have no impact on the union wage.

5. The Bargaining Approach

Next consider more sophisticated methods of reconciling the different preferences of union and management, using a bargaining approach. It turns out that the monopoly union model is nested as a special case within a simple bargaining framework. While some researchers argue that the monopoly union model is a reasonable approximation for particular industries at particular times, observation suggests that wages are more often bargained over by unions and firms. When union and firm bargain over wages, they are in a bilateral monopoly situation, in which a single seller confronts a single buyer of labor. A vital element in this bargaining is the ability of each agent to impose costs on the other agent if it refuses to bargain and engages in a strike or lockout. Although strikes are relatively infrequent, this does not mean that the strike or lockout is unimportant. It is the threat to withdraw that counts. What induces the two parties to bargain is the threat of facing costs imposed by the other party.

Since the union and the firm are in a situation of bilateral monopoly, additional structure must be imposed on the model to generate a determinate solution of the division of the available surplus between the two parties. There are two broad approaches to modeling bargaining behavior—the axiomatic or Nash bargaining approach, and the more recent game theoretical approach (Nash 1950). Under certain conditions, these generate identical predictions in spite of the fact that they proceed from very different assumptions (Binmore et al. 1986).

The solution concept widely used in the union literature is the generalized Nash bargaining solution, according to which wages are determined by maximization of the product of each agent’s gains from reaching a bargain, weighted by their respective bargaining strengths. Since management is assumed to retain the right to determine employment unilaterally, this is referred to frequently as the ‘right-to-manage’ model. Define a status quo or fall-back point for each party as its position if no bargain is reached. For the firm, the status quo is zero, because if it does not reach a bargain with the union workforce, it cannot obtain any other workers since there are no nonunion workers in the sector. So the firm’s net gain is simply its profits function. For the union, its status quo for a representative member is u(b), since that is what a member will receive if no bargain is reached. Thus, the net gain to the union is simply the expected utility of a representative member (see Eqn. (2)) less fall-back utility, given by [u(w) – u(b)]n/t.

The generalized Nash bargaining solution (the product of the weighted net gains to each party) can be written as:

Economic Behavior Of Trade Unions Research Paper

where β measures the bargaining strength of the union and (1 – β) is the bargaining strength of the firm. The solution to this bargaining problem is obtained from the first order condition of Eqn. (6), and is given by:

Economic Behavior Of Trade Unions Research Paper

where ε is the wage elasticity of labor demand. The left-hand side of Eqn. (7) represents the proportional marginal benefit to the bargain from a proportional increase in the wage. Since only the union receives this benefit, it is weighted by the union’s bargaining power β. The first term on the right-hand side is the union’s proportional marginal cost (the percentage reduction in employment due to the proportionate wage increase) weighted by union power β. The second term on the right-hand side represents the firm’s proportional marginal cost, weighted by the firm’s power (1 – β). If the union has all the power, β = 1 and the union extracts the entire surplus—the right to manage model collapses to the monopoly union model. For 0 < β <1, the employment and wage outcome will lie on the labor demand curve anywhere between the monopoly union extreme and the competitive outcome. If the firm has all the power, β = 0 and the wage is then simply the alternative wage b.

The empirically testable predictions of the right-to-manage model are the same as a monopoly union model with the addition of the prediction that variables proxying union strength will affect wages directly.

Is the right-to-manage model Pareto-efficient? (Pareto efficiency is defined by economists as a situation where no one can be made better off without making someone else worse off.) Or can at least one party be made better off by shifting to some other (w, n) pair? In Fig. 2, the isoprofit curves IP represent different levels of profits for the firm, with higher curves denoting lower profits since they are associated with higher wages. IC denotes the union indifference curves, with expected utility increasing as the union moves northeast. Suppose the equilibrium of the rightto-manage model is at point A. The union could be made better off while keeping the firm on the same isoprofit curve, by shifting to point B. Alternatively, the firm could shift to a higher isoprofit curve, while keeping the union on the original indifference curve, by moving to point C. Both parties could be made better off by shifting from A to any point in the shaded area. Positions on the labor demand curve are thus Pareto-inefficient. Either the union or the firm could be made better off by bargaining over employment as well as wages, allowing them to shift from a point such as A to B or C or anywhere in the shaded area. Efficiency in the context of this model is where the marginal rates of substitution of employment for wages, for both union and firm, are equal.

6. The Efficient Bargaining Model

Now suppose the union and firm simultaneously determine wages and employment. While evidence suggests that unions and firms do typically not bargain over wages and employment, this approach has been important in the union literature because of its efficiency property beloved by economists. But the fact that this does not appear to happen in practice suggests that there may be something economists are missing in their modeling. (Of course the wage and employment outcome predicted by the efficient bargaining model is inefficient from the viewpoint of society as a whole.)

Efficiency requires only that one party’s welfare be maximized subject to an arbitrarily fixed level of welfare of the other. This makes it possible to characterize an efficient outcome simply by considering union and firm preferences, without imposing a particular bargaining structure. While this does not yield a unique outcome, it illustrates the necessary conditions for efficiency. An efficient bargain over (w, n) solves:

Economic Behavior Of Trade Unions Research Paper

From the first-order conditions is obtained the result that an efficient (w, n) pair is given by equating the marginal rates of substitution of employment and wages for the union and the firm. This means equating the slopes of the isoprofit and union indifference curves, as illustrated as point C in Fig. 2, and shown below:

Economic Behavior Of Trade Unions Research Paper

The contract curve is defined as the locus of all points satisfying Eqn. (9). Two points on the contract curve are B and C in Fig. 2. The contract curve is positively sloped for risk-averse workers (but vertical if workers are risk-neutral). From Eqn. (9), note that the value of the marginal product of labor pq`(n) is less than the wage rate, by an amount that is equal to the marginal rate of substitution of employment for wages. Hence, the contract curve lies to the right of the labor demand curve for w > b. If w = b, then from Eqn. (9) pq`(n) = w, the perfectly competitive result. Thus, the contract curve and the labor demand curve coincide where w = b. Since for efficiency we only require that one party’s welfare be maximized subject to any arbitrarily fixed level of welfare of the other, any point on the contract curve represents an efficient outcome for a particular arbitrarily fixed level of welfare of one agent.

To determine a unique (w, n) pair, a means of obtaining a particular solution to the bargaining procedure must be added to the model. A common approach is to use the generalized Nash bargaining model of Eqn. (6) modified to allow for bargaining over both w and n. The solution to this maximization problem generates a determinate outcome for wages and employment, which lies somewhere on the contract curve (Leontief 1946, McDonald and Solow 1981). In the 1980s considerable attention was devoted to devising ways of testing which model was appropriate for the union sector (Pencavel 1991).

Many unions are unable to bargain over employment directly, and thus the relevance of this fully efficient model might be questioned. However, unions can often affect employment indirectly, for example, through bargaining over the labor-to-capital ratio, work intensity, or redundancy payments. Johnson (1990) find that, in models in which there is bargaining over wages and the labor-to-capital ratio or work intensity, the wage-employment outcome lies somewhere between the fully efficient and the right-tomanage model.

7. Conclusions

This research paper can only give a flavor of the economics behavior of the trade union, and many important topics have been omitted, for example, the impact of unionization on the macroeconomy (Layard et al. 1991). Other important topics omitted (but covered in Booth 1995) include the economics of strikes, the relationship between unions and hours of work, and the modeling of endogenous union membership, an important topic in countries such as the UK, in which compulsory union membership is illegal. There is also a vast empirical literature estimating the impact of unions on economic performance.

What areas are currently being investigated by economists working on unions? Some recent developments include combining ideas about oligopolistic firm behavior with a unionized labor market in a dynamic framework. While it is too early to see where this literature is heading, this approach might possibly be instructive about appropriate forms of collective bargaining structure. Some countries (such as many European economies) have centralized wage bargaining arrangements, while others such as the US are decentralized. It is important that these institutional differences are incorporated into the modeling frame-work. Calmfors and Driffill (1988) model the impact of collective bargaining structure on economic performance, while Horn and Wolinsky (1988) model choice of collective bargaining structure. These approaches could in principle provide pointers as to what an ‘ideal’ collective bargaining structure might be.

While union membership and power have declined in many industrialized countries in the 1980s and 1990s (for example, Britain and the US), it is unlikely that unions will vanish all together. Indeed, in the future unions may well become stronger in Eastern European, Asian, and African countries. It is therefore important for researchers working on the economics of trade unions to consider what form of unionism might be least distortionary in a modern economy. Moreover, while many models of union behavior refer to explicit labor contracts between a union and management, the models are also relevant to a wider class of situations than those in which a union explicitly represents workers. The models are potentially applicable, with only a few minor modifications, to nonunion labor markets where incumbent workers have some market power and where the firm has economic rents that might be shared.


  1. Binmore K, Rubinstein A, Wolinsky A 1986 The Nash bargaining solution in economic modeling. RAND Journal of Economics 17(2): 176–88
  2. Blair D H, Crawford D L 1984 Labor union objectives and collective bargaining. Quarterly Journal of Economics 99(3): 547–66
  3. Booth A L 1985 The free rider problem and a social custom theory of trade union membership. Quarterly Journal of Economics 100(Feb): 253–61
  4. Booth A L 1995 The Economics of the Trade Union. Cambridge University Press, Cambridge, UK
  5. Bowles S 1985 The production process in a competitive economy: Walrasian, Neo-Hobbesian, and Marxian models. American Economic Review 75(1): 16–36
  6. Calmfors L, Driffill J 1988 Bargaining structure, corporatism and macroeconomic performance. Economic Policy 6: 13–62
  7. Dowrick S 1989 Union-oligopoly bargaining. Economic Journal 99: 1123–42
  8. Faith R L, Reid J D 1987 An agency theory of unionism. Journal of Economic Behavior and Organization 8: 39–60
  9. Farber H S 1978 Individual preferences and union wage determination: The case of the united mine workers. Journal of Political Economy 86(5): 932–42
  10. Freeman R B, Medoff J L 1984 What Do Unions Do? Basic Books, New York
  11. Grout P A 1984 Investment and wages in the absence of binding contracts. Econometrica 52(2): 449–60
  12. Horn H, Wolinsky A 1988 Worker substitutability and patterns of unionisation. Economic Journal 98: 484–97
  13. Johnson G E 1990 Work rules, featherbedding and paretooptimal union–management bargaining. Journal of Labor Economics 8: S237–59
  14. Layard R, Nickell S J, Jackman R 1991 Unemployment; Macroeconomic Performance and the Labor Market. Oxford University Press, Oxford, UK
  15. Leontief W 1946 The pure theory of the guaranteed annual wage contract. Journal of Political Economy 54(1): 76–9
  16. Lindbeck A, Snower D J 1988 The Insider–Outsider Theory of Employment and Unemployment. MIT, Cambridge, MA
  17. McDonald I M, Solow R M 1981 Wage bargaining and employment. American Economic Review 71: 896–908
  18. Nash J F 1950 The bargaining problem. Econometrica 18(2): 155–62
  19. Olson M 1965 The Logic of Collective Action. Harvard University Press, Cambridge, MA
  20. Osborne M J 1984 Capitalist–worker conflict and involuntary unemployment. Review of Economic Studies 51: 111–27
  21. Pencavel J 1991 Labor Markets under Trade Unionism. Blackwell, Oxford, UK
  22. Stewart M B 1990 Union wage differentials, product market influences and the division of rents. Economic Journal 100: 1122–37
Traditions in Sociology Research Paper
Geography Of Tourism Research Paper


Always on-time


100% Confidentiality
Special offer! Get discount 10% for the first order. Promo code: cd1a428655