There has long been a fundamental conflict between labour law and competition or anti-trust law. Labour law supports the policy that workers should have the right to eliminate competition for wages, hours and working conditions through the formation of trade unions and collective bargaining. On the other hand, competition law is based on the idea that free competition between atomistic suppliers of goods and services will result in lower prices to consumers and greater levels of efficiency. In recognition of the conflict between labour law and competition law, legislatures have sought to exempt agreements concerning employment conditions from the application of competition laws. These exemptions have generally applied to both trade unions and employers who seek to act collectively in the industrial context.
However, there would appear to be a number of cogent reasons for reconsidering this traditional assumption as to the dichotomy between labour law and competition law. The reintroduction of sections 45D and 45E, and the introduction of section 45DA, 45DE and 45DB, into the TPA in 1997 undoubtedly make competition law a much more important consideration for trade unions. Also the decision by the Rudd Labour Government not to remove these provisions from the purview of the ACCC, as usually happens when Federal Labour Governments are elected, makes these provisions of continuing relevance for the labour movement.
On a broader level, the rhetoric of competition and efficiency has been the catchcry of many governments in their justification for labour market deregulation, and the Howard Governments emphasis on individual contracts of employment. Indeed, as stated by Hendy and Walton, the rise of individual contracts of employment as the means to achieve a more competitive and efficient labour market has been “the most contagious and virulent industrial relations disease of the late twentieth century in the common law jurisdictions". Indeed the first signs of this "disease" can be seen in the provisions of me Workplace Relations Act 1996 (Cth) ("WRA"), particularly Part VID which creates Australian Workplace Agreements ("AWA’s"),
From a competition law perspective the major issue is the extent to which legislators in common law jurisdictions are seeking to use the rhetoric of competition law and policy to justify the development of a more atomistic labour market through individual contracts. This is being done whilst at the same time ignoring the realities of market dysfunction and inequality of bargaining power which already exist in the labour market, and which will no doubt become more pronounced with the decline of collective bargaining. Ironically, at the same time that legislators seek to deregulate labour markets, they are also recognising that market dysfunction and inequality of bargaining power between businesses is a very real issue in the product market and introducing remedial legislation to combat these problems.
In the following, I will be arguing that labour law and particularly trade unions must seriously consider the relevance of competition law to both their activities and to labour law more generally. In this regard, I will be comparing the development of labour law antitrust jurisprudence in the United States under the Sherman Act, Clayton Act and Norris LaGuardia Act, with recent Australian developments. Of particular importance, is the extension by the US Supreme Court of the labour law antitrust exemption to employers in Brown's case.
I will be arguing that in the context of increased aggregations of employer economic power and the trend to decreased aggregations of labour power, due to the rise of individual contracts, it is imperative for trade unions to challenge the maintenance of a competition law exemption for employers which seek to engage in collective action in labour relations. Indeed, when one examines the policy reasons for an antitrust labour exemption, it becomes readily apparent that mere is little justification for permitting employers the benefit of this exemption.
Conflict between labour law and competition law
The conflict between labour law and competition law arises from the respective treatment of collective action. As stated by Kinter:
Before the rise of labour unions, when individual labourers sold their services to employers, because of the unevenness of bargaining power, employers could play employees off against one another, driving down the wages that were paid. By agreeing collectively to withhold their services, organised labour can enhance its bargaining power and raise the wage level...On the other hand, the heart of antitrust policy is that suppliers of an input should not be allowed to exercise oligopoly power over the prices of goods and services they sell, and thus competition among suppliers is mandated.
Though the above quotation explains the differing treatment accorded to collective action under labour law and competition law, it does not explain the economic factors at play. Competition policy does not advocate competition between sellers as an end in itself; rather competition is a means of achieving the most efficiency use of scarce resources to enhance consumer welfare. Monopoly and oligopoly are seen as a market dysfunction that leads to the misallocation of resources and a net loss to consumers (welfare loss).
In labour economics, the main issue therefore is whether collective action may lead to a misallocation of labour resources and a corresponding welfare loss to society. If collective action results in employees being paid a greater amount than their labour is worth in economic terms, then society as a whole will suffer.
The main reason for requiring collective labour action is the existence of unemployment. Basic market theory holds than the price of a good or service will rise and fall accordingly to demand and supply conditions. In conditions of full employment, labour would is a scarce resource and employers are forced to compete for labour by offering higher wages. On the other hand, with relatively high levels of unemployment employers are able to force wages down due to me availability of replacement employees. Therefore, by definition collective labour action in times of high employment is likely to lead to wage outcomes above market levels. In these circumstances, collective labour action must be justified on equity issues – ie without collective action, employers would be able to both force wages down below subsistence levels and reduce working conditions to unsafe and unhealthy levels.
A further important point is that there is a clear analogy between collective labour organisations and collective capital organisations. A trade union or collective bargaining unit is simply an aggregation of employees into a larger grouping for the purposes of bargaining and protecting their interests; in other words, an organisation for the advancement of collective interests.
Similarly, joint stock companies are simply an aggregation of the capital of individual investors for the purposes of advancing their collective business interests. Competition law does not seek to challenge the aggregations of capital in joint stock companies as "a combination in restraint of trade" or cartel. Rather competition law is only concerned with instances where the level of concentration in an industry increases due to the merger of previously competing companies or where companies seek to use their economic power to engage in anti-competitive conduct.
It should not be thought that the application of competition law to labour law would result in aggregations of labour being subject to the anti-cartel provisions of the TPA (sections 45A and 4D). Rather competition law would treat existing collective labour units as given and focus only on conduct by aggregations of labour for anticompetitive purposes. The main issue therefore would be to determine how competition law would apply to the conduct of existing aggregations of labour. The most effective way of exploring this point is to consider the development of labour law antitrust jurisprudence in the United States.
Labour law antitrust jurisprudence in the United States
The Sherman Act, enacted in 1890, forms the centrepiece of US antitrust regulation. The main substantive provisions are sections 1 and 2 that provide itfer alia:
Section 1 - Every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations, is hereby declared to be illegal.
Section 2 - Every person who shall monopolise, or attempt to monopolise, or combine or conspire with any other person of persons, to monopolise any part of trade or commerce among the several states, or with foreign nations, shall be deemed guilty of a misdemeanour...
An important point to recognise about the Sherman Act is that the driving force behind its introduction was public concern about aggregations of capital and not aggregations of labour. As stated by Justice Stone in the Apex Hosiery case the Sherman Act was "enacted in era of trusts and of combinations of businesses and of capital organized and directed to control of the market by suppression of competition in the marketing of goods and services, the monopolistic tendency of which had become a matter of public concern".
From the Congressional debates which preceded the enactment of the Sherman Act, it is apparent that the Act that was not intended to apply in any way to labour. Despite the clear Congressional intent that the Sherman Act was to apply to aggregations of capital and not labour, the Courts were quick to use the new provisions of sections 1 and 2 to constrain labour activities; indeed, the Act was applied more often to labour unions than businesses.
The first major case in which the Supreme Court considered the application of the Sherman Act to union activity was the Danbury Hatters case. In this case, the United Hatters of North America sought to force Loewe, a hat manufacturer, to unionise its workforce by inducing a strike at its plant and organising a boycott of its products. In its judgement, the Supreme Court specifically considered the issue of whether the Sherman Act had been intended to apply to labour activities and concluded:
The combination charged falls within the class of restraints of trade aimed at compelling third parties and strangers involuntarily not to engage in the course of trade except on conditions that the combination imposes and there is no doubt that…"at common law every person has individually, and the public also has collectively, a right to require that the course of trade should be kept free from unreasonable obstruction.
The Court argued that the particular combination formed by the United Hatters was the type of combination which was prohibited under the common law as a restraint of trade. As has been argued by Boudin, boycotts were not illegal at common law as a restraint of trade but rather as a private wrong. Accordingly, the basic premise on which the Supreme Court relied to apply the Sherman Act to labour boycotts was flawed.
The Court also considered the issue of Congressional intent. In this regard, the Court relied on the Amalgamated Council case:
I think the congressional debates show mat me statute had its origin in the evils of massed capital; but when Congress came to formulating the prohibition, which is the yardstick for measuring the complainant's right to the injunction, it expressed it in these words: 'Every contract or combination in the form of trust or otherwise in restraint of trade or commerce among the several states or with foreign nations, is hereby declared to be illegal'. The subject had so broadened in the minds of the legislature that the source of the evil was not regarded as material, and the evil in its entirely is dealt with. They made me interdiction (to) include combinations of labor as well as of capital; in fact, all combinations in restraint of commerce, without reference to the character of the persons who entered into them.
The Supreme Court simply accepted the above comments from Amalgamated without actually checking the Congressional debates. Had they done so it would have become apparent that Congress had absolutely no intention of applying the Sherman Act to labour activity.
As a result of these two false premises, namely that boycotts were illegal at common law as a restraint of trade and that Congress intended the Sherman Act to apply to "combinations of labor", the entire range of union action became subject to antitrust challenge, a position which the Supreme Court has never completely resiled from to this day.
After '"unceasing agitation" by labour to have the Sherman Act amended to provide an exemption for particular labour activities, the Clayton Act was passed in 1914. Section 6 provided a broad exemption for trade union action as follows:
That the labor of a human being is not a commodity or article of commerce. Nothing contained in the antitrust laws shall be construed to forbid the existence and operation of labor, agricultural, or horticultural organisations, instituted for the purposes of mutual help, and not having capital stock or conducted for profit, or to forbid or retain individual members of such organisations from lawfully carrying out legitimate objects thereof; nor shall such organizations, or the members thereof, be held or construed to be illegal combinations or conspiracies in restraint of trade, under the antitrust laws.
Section 20 prohibited the seeking of injunctions under the Sherman Act for listed labour activities, including the termination of any relation of employment ceasing to perform any work or labour, or persuading others through peaceful means to abstain from working. Section 20 also sought to limit access to injunctions under the Sherman Act in circumstances where remedies were available under other laws.
The significance of sections 6 and 20 of the Clayton Act for trade union action cannot be underestimated. Not only did section 6 recognise the existence and operation of trade unions, but it also recognised explicitly the special status of labour which justified its exemption from antitrust laws; namely that the "labor of a human being is not a commodity or article of commerce”.
One would have been expected that the combined effect of sections 6 and 20 would provide labour with a complete exemption from the application of the antitrust laws. However, it soon became apparent that the Courts were to construe these provisions very narrowly.
In the Duplex case, the Supreme Court focused on the use of the words lawfully and legitimate in section 6 of the Clayton Act in their consideration of a boycott engaged in by the International Association of Machinists to force an employer to unionise. The Court concluded that whilst section 6 provides trade unions with some protection, there was no Congressional intent to permit "unlawful acts” for "illegitimate ends". The Court also interpreted section 20 as being limited to the employee - employer relationship; therefore where third parties were the subject of boycott conduct, section 20 could not be relied upon to protect the union."
After further agitation by labour, the Norris LaGuardia Act was passed in 1932 to circumscribe the circumstances where an injunction may be granted against labour activities under the Sherman Act. The impact of this Act in substantive terms was to restate the exemptions previously provided under the Clayton Act with greater specificity and to expand those categories. Federal Courts were prevented from granting injunctions upon the ground that any of the persons participating or interested in a labour dispute constitute or are engaged in an unlawful combination or conspiracy because of the doing in concert of the acts enumerated in section 4, and to remove the requirement of an employee - employer relationship.
However of central importance is the clear statement of legislative policy concerning the application of antitrust laws to labour contained in section 2 of the Norris LaGuardia Act:
Whereas under prevailing economic conditions, developed with the aid of governmental authority for owners of property to organise in the corporate and other forms of ownership association, the individual unorganized worker is commonly helpless to exercise actual liberty of contract and to protect his freedom or labor, and thereby to obtain acceptable terms and conditions of employment, wherefore, though he should be free to decline to associate with his fellows, it is necessary that he have full freedom of association, self-organization, and designation or representatives of his own choosing, to negotiate the terms and conditions of his employment, and that he shall be free from the interference, restraint, or coercion of employers of labor, or their agents, in the designation of such representatives or in self-organisation or in other concerted activities for the purpose of collective bargaining or other mutual aid or protection...
The above statement of Congressional intent is notable in its recognition of the inability of many workers to safeguard their interests against employers due to their lack of bargaining power and further that the notion of freedom of contract is largely a myth in the employment relationship. The statement is also notable in its recognition that collective bargaining is a necessary means of redressing the inequality of bargaining power which exists between employer and employee.
The implications of the opening phrase are also of particular relevance to an understanding of the relationship between competition law and labour law. As is apparent, Congress specifically draws attention to the fact that aggregations of capital in "corporate and other forms of ownership association” are only possible due to "government authority". Therefore rather than see aggregations of labour into trade unions as constituting a special category of combination, Congress explicitly recognised such combinations were both justifiable and necessary as a counterbalance to aggregations of capital.
The changes brought about by the Norris LaGuardia Act were first considered by the Supreme Court in the Apex Hosiery case. In that case, the American Federation of Full Fashion Hosiery Workers took possession of a hosiery plant and held a prolonged "sit-down" strike during which much of the machinery was destroyed in an attempt to force the employer to unionise. Although the Court specifically considered the effect of the Norris LaGuardia Act, they concluded that it did not show a clear Congressional intent to exclude wholly labour from the application of the Sherman Act.
The Supreme Court also took the opportunity in this case to reformulate the underlying purpose of the Sherman Act cases and the test to be applied when considering combinations in restraint of trade:
In the cases considered by this Court since the Standard Oil case in 1911 some form of restraint on commercial competition has been the sine qua non to the condemnation of contracts, combinations or conspiracies under the Sherman Act, and in general restraints upon competition have been condemned only when their purpose or effect was to raise or fix the market price.
Accordingly, the Supreme Court concluded that the Sherman Act did not apply to the unions conduct as it did not have the purpose or effect of interfering with "commercial competition" by raising or fixing the market price of goods.
The establishment in Apex Hosiery of the "purpose and effect” test was the most significant development in labour law antitrust jurisprudence in the United States, and one which has formed the basis for Australian labour prohibitions under the TPA. In formulating this test, the Supreme Court gave primacy not only to the effect of a combination on competition in product markets, but also whether such an anticompetitive effect was the intended consequence of labour activity. As is apparent, the introduction of purpose also had the effect of significantly increasing the evidentiary burden of litigants seeking to use the Sherman Act against labour activities.
Despite the significant advances made in the Apex Hosiery case, it is arguable that the Supreme Court had not yet truly recognised the importance of the Clayton Act and Norris LaGuardia Act in their approach to labour activities under the Sherman Act. Such recognition came in Hutcheson's case, decided a year later. In this case, Justice Frankfurter spelled out, not only the full importance of these statutes, but also the way in which the Court had been ignoring clear Congressional direction. The facts of the case may be briefly summarised as arising out of a demarcation dispute between the United Brotherhood of Carpenters and the International Association of Machinists. The Carpenters demanded work traditionally undertaken by the Machinists. When Anheuser-Busch, the employer, refused the demand, the Carpenters Union picketed the employer’s business and called for a consumer boycott of its products, Anheuser-Busch beer.
In his judgment, Justice Frankfurter claimed that despite the clear Congressional intent that the application of the Sherman Act to labour was to be circumscribed, the judiciary had subverted this intention by construing the provisions of the Clayton and Norris LaGuardia Acts "in a spirit of mutilating narrowness”. He posited the following as the correct judicial approach to considering trade union conduct:
Whether trade union conduct constitutes a violation of the Sherman Act is to be determined only by reading the Sherman law and sec. 20 of the Clayton Act and the Norris LaGuardia Act as a harmonizing text of outlawry of labour conduct.
In other words, the three Acts established a complete and exclusive set of rules for dealing with trade union conduct under the antitrust laws. It was not permissible for the Court to move outside these boundaries to extend the application of the Sherman Act to labour conduct. Ironically, that it exactly what Frankfurter proceeded to do in his judgment when he stated:
So long as a union acts in its self interest and does not combine with non-labour groups, the licit and illicit under section 20 (Clayton Act) are not to be distinguished by any judgment regarding the wisdom or unwisdom, me lightness or wrongness, me selfishness or unselfishness of the end which the particular union activities are the means.
In this passage, Frankfurter establishes two new criteria for evaluating labour conduct ie: self-interest and combination with non-labour groups. The first criteria, rather than limiting the application of the law introduced a new requirement which not only reinforced the Court’s power to consider the unions reasons for engaging in particular conduct but also to potentially circumscribe the powers of unions to engage in sympathy strikes where it could be argued they were not seeking to advance their own self interest. Frankfurter’s second requirement, namely the prohibition on unions combining with non-labour groups, was justified as an application of the decision in Brims case. A review of this decision shows that the Court did not decide that a combination between a union and a non-labour group was per se subject to Sherman Act consideration.
Although Frankfurter's judgment was a significant step in clarifying the purpose and effect of the antitrust statutes, his extension of union liability in the two areas identified above, provided the judiciary with a relatively wide discretion in considering labour conduct under the Sherman Act.
The most notable application of Justice Frankfurter's "non-labour combination” restriction was the Allen Bradley case. In this case, the International Brotherhood of Electrical Workers secured closed shop agreements with electrical contractors in New York through strikes and boycotts. The union then sought to expand its coverage to electrical manufacturers in New York. Under the union's agreement with the local electrical manufacturers, it guaranteed that New York electrical contractors would be obliged to purchase equipment from only local electrical manufacturers, which operated a closed shop. Manufacturers were also obligated to restrict their sales to local electrical contractors, which operated closed shops. As a result, of these agreements, non-unionised electrical contractors and electrical manufacturers were excluded from the New York market. The Court in its judgement comments on the success of these agreements:
The combination between the three groups, union, contractors, and manufacturers, became highly successful from the standpoint of all of them. The business of the New York City manufacturers had a phenomenal growth, thereby multiplying the jobs available for the Local's members. Wages went up, hours were shortened, and the New York electrical equipment prices soared, to the decided financial profit of local contractors and manufacturers.
The conclusion of the Court was that as the union had participated with non-labour groups for the purpose of eliminating competition in the supply of electrical goods, the conduct was not immune from antitrust attack. The Court also suggested that, to the extent that the union had acted unilaterally, the conduct would have been immune from the Sherman Act.
The decision in this case demonstrates precisely the area where it is legitimate for the antitrust law to apply to labour activities. As stated above, labour law permits trade unions to take the competition out of wages by acting collectively, whether through bargaining or through pickets and strikes. However, where a trade union seeks to assist in the formation of a single firm monopoly, or a cartel or oligopoly in a market, such conduct should be subject to antitrust attack.
Labour is best served by a multiplicity of employers in any given industry as it is only through such competition will wage levels be maximised. On the other hand, unions sometimes believe that the only way to achieve higher wages is to assist an employer gaming a monopoly or dominant position in a market on the assumption that they will be able to appropriate a proportion of the employer's monopoly rents. The belief amongst some unions that there is a benefit in promoting uncompetitive product markets to obtain higher wages is misplaced.
Firstly, by reducing the number of buyers which are competing for labour, such union conduct is more likely to result in lower wage levels in the longer term. Secondly, the belief that labour can appropriate monopoly rents from an employer for any extended period is erroneous, particularly in industries where barriers to new entry are high. In such industries, employers are more likely to use labour to achieve a monopoly position, and then take swift action to depress wage levels on the basis that a monopoly supplier is also a monopoly employer.
Following the decision in Allen Bradley there were few significant antitrust cases involving labour activity until 1965 when Pennington and Jewel Tea were decided. These cases are significant to the extent that they proposed a three-step analysis of labour activities under the Sherman Act.
In Pennington, the United Mineworkers Union had entered into an agreement with a multiemployer bargaining unit which permitted mine mechanisation and redundancies on condition that the remaining workers be paid higher wages. Phillips Brothers, a smaller coal miner, alleged that the UMW had also reached a broader agreement with large coal miner operators to impose a higher wage rate on all employers to drive smaller employers out of the market, reduce coal oversupply and push coal prices up. In this case, Justice White applied the following three-step test to assessing the UMW's liability under the Sherman Act:
1. whether the union was acting alone - this was seen as an application of the test established by Justice Frankfurter in Hutcheson
2. whether the restraint was of a kind forbidden by the antitrust laws - this was seen as an amalgamation of the test in Apex Hosiery (purpose and effect of affecting commercial competition) and Hutcheson (acting in pursuance of own self interest) and
3. whether national labour policy required that the immunity apply notwithstanding an adverse finding in relation to issues (1) and (2) above.
The final step in the Court analysis of the conduct was a radical departure from earlier antitrust cases involving union activities. Despite the passage of the Wagner Act in 1935, considerations of national labour policy had rarely been mentioned in applying the Sherman Act to labour activities. In considering this issue, Justice White stated:
But there is nothing in the labor policy indicating that the union and the employers in one bargaining unit are free to bargain about the wages, hours and working conditions of other bargaining units or the attempt to settle these matters for the entire industry. On the contrary, the duty to bargain unit by unit leads to quite a different conclusion. The union's obligation to its members would seem best served if the union retained the ability to respond to each bargaining situation as the individual circumstances might warrant, without being straight-jacketed by some prior agreement with the favoured employers. 
Therefore, had the conduct of the UMW been permitted under the national labour policy, there may have been grounds for an immunity. As there were no such grounds, the agreement must be characterised as simply a means of facilitating collusion in the coal industry.
The Pennington case was the stage at which US antitrust law and labour law converged. It signalled that in future, to the extent that labour activity was permitted under the national labour policy, it would be immune from antitrust attack.
The Jewel Tea case also involved a multiemployer bargaining unit. The case is less significant than Pennington as there was no probative evidence of a conspiracy between the Amalgamated Meat Cutters and a group of retailers. However, in the course of its judgment the Court made two significant observations about the scope of antitrust immunity for labour activities. Firstly, the Court explained how the national labour policy should be applied:
A provision establishing the particular hours of work would be within the ambit of wages, hours and working conditions requiring mandatory bargaining, and the unions’ success in obtaining that provision through negotiation in pursuit of their own policies falls within the protection, of the national labor policy and is exempt from me Sherman Act.
In other words, in the absence of evidence of combination between the union and non-labour groups, agreements on mandatory bargaining issues under the Wagner Act (such as wages, hours of work and working conditions), should be immune from the Sherman Act. The justification for this exemption was the "strong labour policy favouring the association of employees to eliminate competition over wages and working conditions”.
Secondly, the Court stated that in determining whether a restraint imposed by a union was necessary to protect it own interests, it is relevant to consider whether the agreement was the least restrictive alternative available in terms of its effect on the relevant product market. This test would seem to qualify Justice Frankfurter's test in Apex Hosiery as follows — a union is immune from the Sherman Act as long as it acts in its self-interest, using the least restrictive method available to restrain competition in the relevant product market. As is apparent this is a significant restriction on the scope of the labour immunity, as by definition the most restrictive measures on the final product market are also the most effective measures available to the union.
The final case of particular relevance to the development of labour law antitrust jurisprudence is Brown. In this case, the Supreme Court considered whether an agreement between the National Football League and a number of football clubs to pay pro-football players a particular wage per week contravened the Sherman Act. The agreement came about following multi-employer collective bargaining negotiations had reached an impasse.
After considered the scope of the labour exemption from antitrust law, the Court concluded that the employers should be able to take advantage of this exemption on the basis that:
The post-impasse imposition of a proposed employment term concerning a mandatory subject of bargaining is unobjectionable as a matter of law and policy, and, indeed, plays a significant role in the multi-employer collective bargaining process that itself comprises an important part of the Nation's industrial relations system. Subjecting it to antitrust law would threaten to introduce instability and uncertainty into the collective bargaining process, for antitrust often discourages the kinds of joint discussions and behaviour that collective bargaining invites or requires.
In Brown the Supreme Court extended the application of the antitrust exception established in Pennington and Jewel Tea to cover employer organisations which may combine to agree on the wages it will offer employees on the basis that it is necessary to ensure the stability of the collective bargaining system as a whole.
The main problems with this extension of the exemption were outlined in the dissenting opinion delivered by Justice Stevens. The most obvious objection to the Brown decision is that it ignores the reasons why the antitrust exemption was specifically conferred on labour, rather than capital; namely the recognition of the significant inequality of bargaining power that exists between employees and employers. As stated by Justice Stevens:
…neither the policies underlying the two separate statutory schemes, nor the narrower focus on the purpose of the non-statutory exemption, provides a justification for exemption from antitrust scrutiny collective action initiated by employers to depress wages below the level that would be produced in a free market.
The decision in Brown creates a wide range of possibilities for employers in multi-employer bargaining units. Furthermore, it is not clear whether collective agreements by employers outside such units may be able to take advantage of this exception. To the extent that the decision expands the ability of employers to determine collectively wage levels, it is likely to raise some concerns for trade unions
Australian labour law exemption from competition law
Section 51 (2)(a) of the TPA establishes a broad exemption from competition law:
In determining whether a contravention of a provision of (his Part (Part IV) other that section 45D, 45DA, 45DB, 45E, 45EA or 48 has been committed, regard shall not be had:
(a)to any act done in relation to, or to the making of a contract or arrangement or the entering into of an understanding, or to the provision of a contract, arrangement or understanding, to the extent that the contract, arrangement or understanding, or the provision, relates to, the remuneration, conditions of employment, hours of work or working conditions of employees.
The main effect of the section is to exempt specific categories of contracts, arrangements and understandings from being contraventions of Part IV; namely –
• Section 45 — contracts, arrangements or understandings that restrict dealings or affect competition, including price fixing under section 45A and market sharing under section 4D
• Section 45B - covenants affecting competition
• Section 45C- covenants in relation to prices
• Sections 46 and 46A -misuse of market power and
•. Section 47 - exclusive dealing.
These exclusions equate broadly to the types of conduct that are prohibited under section 1 and 2 of the Sherman Act.
The exemption does not apply to the primary and secondary boycott provisions of the TPA (sections 45D, 45DA, 45DB, 45E and 45EA) nor somewhat surprisingly to section 48 which prohibits resale price maintenance.
It should also be noted that section 51 (2)(a) exempts any party who enters into a contract, arrangement or understanding which relates to remuneration, conditions of employment, hours or work or working conditions. The exemption is available to both employers and employees.
Assessment of section 51(2)(a)
Comparing the approach of the Australian Parliament in providing a labour exemption under section 51(2)(a) with that of the US Congress a number of significant differences can be identified.
Firstly the Australian Parliament decided to draw a broad exemption under section 51(2)(a) for conduct which may be characterised as legitimate union conduct. The TPA then identifies specific categories of conduct that would constitute a contravention – ie sections 45D, 45DA, 45DB, 45E and 45EA. This contrasts with the approach of Congress in both the Clayton and Norris LaGuardia Acts which opted to create narrow exemptions based on specific types of union conduct.
Secondly, it is also notable that the listing of the remuneration, conditions of employment, hours of work or working conditions of employees in section 51(2)(a) could be seen to accord with the areas identified in Pennington and Jewel Tea as requiring exemption from the antitrust laws on the basis of the national labour policy.
Thirdly, there is no reference in section 51 (2)(a) to considerations such as whether the union has combined with non-labour groups to bring about a restraint of trade or whether the union’s purpose and effect was to effect commercial competition. As will be discussed below, these issues are highly relevant to the application of the primary and secondary boycott provisions of the TPA.
Finally, the TPA exemption applies to both employee and employer conduct in forming a contract, arrangement or understanding, which is similar to the position in the US since Brown.
There are a number of similarities and differences between the labour law antitrust exemption under US law and Australian law. The most important difference is that whilst Congress established a broad prohibition on restraints of trade, which was interpreted by the Courts to apply to most forms of labour activity, the Australian Parliament sought to establish a broad exemption. This is largely the reason why there have been so few cases in Australia on the scope of the exemption, and apparently just as little debate as to the desirability of having such a broad exemption.
TPA prohibitions on labour conduct
The main labour prohibitions are contained in sections 45D, 45DA, 45DB, 45E and 45EA. Although I do not propose to discuss these provisions in any detail, some general observations are necessary to draw meaningful comparisons between Australian and US prohibitions on labour conduct
Section 45D is a prohibition on secondary boycott conduct. It applies to all persons seeking to act collectively to interfere with the performance of a contract between a third and fourth person. The fact that the section is in reality directed primarily to unions is demonstrated by subsections 45D(3) and (4) which limit the application to situations where the third or fourth person are corporations. In the US, such secondary boycott is prohibited both under the Wagner Act and also under State laws. Secondary boycott may also be subject to action under the Sherman Act where the three-step test established in Pennington is satisfied.
As is apparent most cases under section 45D would not satisfy the second limb of Pennington which requires that the conduct engaged in by the union have the purpose and effect of affecting commercial competition. The standard applied under section 45D is limited to the "purpose and effect of causing substantial loss of damage to the business of the fourth person". In this regard, section 45D is not aimed at advancing antitrust goals.
Section 45E prohibits persons, primarily employers, from entering into agreements with unions which have the effect of preventing the supply or acquisition of goods of services. In essence, section 45E aims at prohibiting employers agreeing to union demands that they not deal with another business. To the extent that section 45E prohibits unions combining with non-labour, it is analogous to the rule established by Frankfurter in Apex Hosiery. It is also apparent that primary liability for a breach of section 45E is incurred by the employer and not the union. As was the case with section 45D, the effect of the conduct on competition is not a relevant consideration in the application of section 45E.
Section 45DB is not aimed at prohibiting restrictions on competition. Rather the section prohibits persons acting in concert to hinder or prevent the movement of goods between Australia and places outside Australia. This section is also clearly the widest of the three non-competition provisions in the TPA, as it prohibits the preventing or substantial hindering of international trade.
Section 45DA prohibits secondary boycotts undertaken for the purpose, and which have the effect, of substantially lessening competition in a market. The section is also subject to a range of defences in section 45DD, including when the dominant purpose of the union's conduct relates to employment matters. Accordingly, this provision can be seen as similar to the Pennington test in that the purpose and effect of the union's conduct must be to substantially lessen competition in a product market, and there is a defence under section 45DD where the union’s actions relate to matters of legitimate self-interest
Finally section 45EA, extends the prohibition in section 45E to conduct which is engaged in with the purpose, and which is likely to have the effect, of substantially lessening competition in a market, Again as the primary contravener of the provision would be the employer, it could not be considered a direct prohibition on union activity. In this regard, the TPA differs from the interpretation of the Sherman Act. Whilst the TPA puts primary liability for a union/non-labour combination under section 45EA on the non-labour participant, the Sherman Act would impose liability both the union and non-labour participants.
In summary, of the above prohibitions only section 45D, 45DA and 45DB are direct prohibitions on labour activities. Of these provisions, only conduct prohibited under section 45DA is prohibited under the Sherman Act. This is because section 45DA is the only labour activity prohibition in the TPA that is concerned with the effect of union action on competition in a product market. One can conclude that the application of the TPA to labour activities is much narrower than the Sherman Act, at least in terms of prohibitions on anticompetitive conduct by unions.
To the extent that the TPA includes prohibitions on secondary boycott conduct which raise no competition issues, it is considerably broader than the Sherman Act. As stated above, in the US such secondary boycott conduct would be dealt with under the Wagner Act or State legislation.
Extension of reach of competition law to labour activities
In 1999, the NCC conducted a review of the section 51(2)(a) exemption. In this regard it issued a draft report recommending the retention of the exemption on the basis that:
The objectives of section 51(2)(a) is to excise the labour market from goods and services markets for the purpose of applying competition law, supporting a public policy that labour markets should be regulated through the industrial relations framework and not the TPA.
The NCC believed that that labour market should be “excised" from competition law on public policy grounds.
The DEWRSB however took a different view. In its submission to the NCC it stated:
While recognising me difficulty of reconciling the competing policy objectives of labour and competition law, DEWRSB is of the view that a case exists for narrowing the scope of the exemption under section 51(2)(a). In particular, amendments could be made to establish arrangements whereby an exemption like that provided under section 51(2)(a) remains in place. But is revocable where the anti-competitive detriment of the particular arrangements in question outweighed the labour law policy objectives.
As is apparent DEWRSB conceded that the application of competition law to the labour market involves reconciling competing policy objectives. The Department was not suggesting the removal of the section 51(2)(a) exemption altogether, rather it was proposing that the exemption be revocable in certain circumstances.
The main flaw with the DEWRSB proposal was that it did not provide any detail of the circumstances where it would have been appropriate to revoke the exemption due to the “anti-competitive detriment” of an arrangement. Accordingly, conferring on the ACCC of a broad discretion to review labour contracts, arrangements and understandings would have created a degree of uncertainty in labour contract negotiations, which in turn would have been likely to undermine the collective bargaining system.
Most types of labour conduct which has been prohibited under the Sherman Act since Apex Hosiery are already subject to the TPA. In most instances where a union either seeks to combine with non-labour groups to prevent an employer from entering a market or acts unilaterally to prevent an employer from entering a market, the TPA will is adequate to deal with the conduct under section 45 or even section 46, despite section 51(2)(a) exemption. A price fixing agreement between a union and a number of employers is likely to be subject to section 45, through the operation of section 45A, where it has the effect of "fixing, controlling or maintaining prices” in the final product market
Section 47 could apply to union conduct, in a situation where the union insisted on an agreement with an employer that it would not hire non-union members. While such an agreement could be challenged under section 47, such agreements are also likely to contravene the freedom of association provisions contained in the WRA.
The TPA is already an effective means of dealing with the types of union conduct which have been identified by the US Supreme Court as being subject to the Sherman Act. In these circumstances, There is little benefit in seeking to extend the scope of the TPA to deal with labour conduct, unless specific examples of conduct can be identified which falls outside the current scope of the TPA.
An area of significant concern for Australian trade unions should be the retention of the employer exemption under section 51(2)(a). It has been argued that the retention of the exemption is based largely on Pennington type grounds, namely to preserve the national labour policy. However, the uncritical acceptance of this position may be detrimental to trade unions.
The driving ideological assumptions behind labour market reform are the market goals of efficiency and competitiveness. For example, as stated by John Howard in his speech "A Competitive Australia" in 1995 he stated:
For too long trade union bureaucracies have thrived on the belief that without compulsion, a legislative monopoly and the conveniently belong provision, the union movement is doomed. This is a highly defensive and defeatist approach...
Just as we have removed statutory protection from various industries and activities, the same must apply to the trade union movement. Unions are ultimately a component of me service industry. They compete in the market for labour representation, and they must be subject to market disciplines.
The enactment of the WRA represented the first stage in the attempt to realise the broader ideological goal of a highly deregulated labour market. Furthermore an essential prerequisite to achieving this goal was to break down both the power of trade unions and the scope of collective bargaining. In this regard it is relevant that the WRA sought to break down alleged union monopolies through a number of mechanisms including:
* the revised conveniently belong rule
* enterprise based associations
* freedom of association provisions and
* disamalgamation provisions.
Also the introduction of Australian Workplace Agreements, was a measure that was clearly directed to advancing individualistic goals of freedom of contract to the detriment of collective action. Although the changes brought about by the WRA did not go as far as me Coalition Government has initially proposed, it was likely that further changes were to be introduced.
The changes brought about by the WRA had the effect of increasing the inequalities of bargaining power that existed between employers and employees. Bargaining power between employers and employees can only be equalised where there is strong trade union organisation and a comprehensive system of collective bargaining.
The fundamental irony is that while the WRA sought to promote the goals competition and efficiency by advocating freedom of contract in the labour market, the TPA was amended at the same time to reflect the realisation that freedom of contract does not necessarily exist in many product markets. In 1998, the Federal Government introduced new legislation providing relief for small businesses on the grounds of unconscionable conduct; sections 51AC and 51AD. This legislation was enacted on the basis than in many circumstances small businesses were in the same position as consumers in their dealings with larger suppliers, ie effectively powerless. Accordingly, as the inequality of bargaining power between small business and large suppliers was so great, the TPA needed to be amended to provide relief.
In the context of reducing trade union power and more atomised labour markets, the retention of the employee exemption under section 51(2)(a) is vital in at least in a symbolic sense to avoid possible further erosion of the notion of collective bargaining.
The employer exemption from competition law is unjustifiable and potentially damaging for labour. As argued above in the consideration of US antitrust law, the exemption of labour activities from antitrust laws was based on the inequalities of bargaining power between employers and employees. Accordingly, there was no justification for extending the protection to employers. More importantly, the employer exemption was likely to become much more valuable to employers in the future as the labour market becomes more atomised. If workers are to ensure maximum competition amongst employers for their labour, it is imperative that the exemption for collective action between employers be removed.
Whilst a fundamental conflict between labour law and competition or anti-trust law has long existed, there is a significant interaction between the two as shown by the development of US jurisprudence. It appears appropriate for a range of labour activities to be subject to scrutiny under antitrust/competition laws. In this regard, the US legislation and case law provides a useful analytical guide to Australian legislators and practitioners on the types of issues which are relevant in assessing the competitive effects of particular types of labour conduct
Adequate enforcement remedies already exist under the TPA to deal with the types of conduct identified by the US Supreme Court as properly the subject of the antitrust laws. Indeed the scope for action under the TPA in relation to labour activities is considerably broader than the scope of action permitted under the Sherman Act due to the inclusion of the number of primary and secondary non-competition prohibitions.
The issue of central concern is the way in which deregulation of the labour market was being promoted during the Howard Government using the rhetoric of competition and efficiency, with little regard to the market structures which will invariably result. Freedom of contract will not achieve the best outcomes in the labour market for precisely the same reasons that freedom of contract has not achieved the best outcomes in product markets - disparities in the bargaining power between buyers and sellers. Indeed, in Australia the existence of such disparities of bargaining power in product markets were seen as so serious that legislative intervention was deemed necessary.
It is trite to say that significant disparities of bargaining power exist in labour markets; the more valuable observation is that such disparities are likely to become much more pronounced in the future with the state sponsored decline of trade unions and collective bargaining and the rise of an atomistic labour market through individual contracts of employment. This concern remains with the Rudd Labour Government proposing to retain AWA's to some degree.
A factor which has the potential to add to this disparity of bargaining power is the retention of the competition law exemption for collective employer action on the terms and conditions of employment. It is imperative that trade unions challenge the retention of this exemption to as to prevent the creation of legally sanctioned cartels of labour buyers.
 See J Meltzer, "Labour Unions, Collective Bargaining and the Antitrust Laws" (1965) 32 The University of Chicago Law Review 659; R K Winter, “Collective Bargaining and Competition: The Application of Antitrust Standards to Union Activities' (1963) 73 The Yale Law Journal 14; H Shuhman, "Labor and the Antitrust Laws' (1940) 34 Illinois Law Review of Northwestern University 769; H N Bernhardt, "The Allen Bradley Doctrine: An Accommodation of Conflicting Policies" (1962) 110 University of Pennsylvania Law Review 1094; and A Cox, "Labor and Antitrust Laws - A Preliminary Analysis" (1955) 104 University of Pennsylvania Law Review 254.
 See D Chin, “Exhuming the Individual Employment Contract: A Case of Labour Law Exceptionalism" (1997) 10 Australian Journal of Labour Law 257 and R Naughton, "Sailing into Uncharted Seas: The Rote of Unions Under the Workplace Relations Act 1996 (Cth)(1997) 10 Australian Journal of Labour Law 112.
 J Hendy and M Walton, "An Individual Right to Union Representation in International Law" (1977) 26 Industrial Law Journal 205.
 See Chin, op.cit. pp. 265-276. Note the subsequent introduction of section 51AC to the TPA in 1998 which is aimed at addressing the inequality of bargaining power between small businesses and suppliers.
 Brown v Pro Football. Inc. dba Washington Redskins (1996) US No 95-388.
 E W Kinter, "Exemptions for Labour Union Activities and Associated Employee Relations” in E W Kinter, JP Bauer and WP Kratzke, Federal Antitrust Law: A Treatise on the Antitrust Laws of the United Slates, Anderson Publishing, Cincinnati, 1989, p. 286.
 Apex Hosiery Co v Leader 310 US 469 (1940) per Stone J.
 For a detailed commentary of the Congressional debates leading up to the enactment of the Sherman Act see L B Boudin, "The Sherman Act and Labour Disputes: I” (1939) 39 Columbia Law Review, pp. 1281-1337 and LB Boudin, "The Sherman Act and Labour Disputes: II” (1940) 40 Columbia Law Review, pp. 14-51.
 A Cox, S Curtis, R A Gorman, Labor Law: Cases and Materials, The Foundation Press, Inc., Eighth Edition, New York, 1977, p. 36.
 Loewe v Lawlor (1908) 208 U.S. 274. Although the Sherman Act was used to counter boycott action against Pullman Company by the American Railway Union in 1894, the Supreme Court in affirming me earlier decision did not rely on the Sherman Act for its decision - In Re Debs (1895) 158 U.S. 564.
 (1908) 208 U.S. 274; at p. 297.
 Boudin, op. cit., pp. 19-51 (Part II).
 United Stales Workingmen's Amalgamated Council (1993) 54 Fed 994.
 Ibid. p. 996.
 Boudin, op. cit., pp.1286ff (Part 1)
 See A G Siegal, W B Connolly Jr. and R K Walker, "The Antitrust Exemption for Labor - Magna Carta or Carte Blanche?" (1975) 13 Duquesne Law Review 411; at p. 420.
 Duplex Printing Press Co v Deering (1921) 254 U.S. 41.
 Siegal, et. al., op. cit. pp. 427-429.
 Section 4.
 Section 5.
 Section 13(c)
 Apex Hosiery Co v Leader (1940) 310 U.S. 469
 Ibid., p. 493.
 Ibid., p. 485.
 United States v Hutcheson (1940) 312 U.S. 219.
 Ibid., p. 235.
 Ibid., p. 233.
 Ibid., p. 234.
 United States v Brims (1926) 272 U.S. 549.
 Allen Bradley Co. v Local No. 3. IBEW (l945) 325 U.S. 797.
 Ibid., p. 798.
 United Mineworkers v Pennington (1965) 381 U.S. 657.
 Local 189, Amalgamated Meat Cutters & Butcher Workmen of North America. AFL-CIO v Jewel Tea Co. (1965) 381 U.S. 676.
 (1965) 381 U.S. 657; at pp. 664-5.
 Ibid., pp. 665-6; at p. 668.
 Ibid., pp. 666-7.
 See D L Leslie, “Principles of Labor Antitrust" (1980) 66 Virginia Law Review 1183; at pp. 1204-18.
 (1965) 381 U.S. 676; at pp. 689-91.
 Connell Co v Plumbers & Steamfitters (1975) 421 U.S. 616.
 (1965) 381 U.S. 676; at pp. 692-3.
 Brown v Pro Football. Inc. dba Washington Redskins (1996) US No 95-388.
 Ibid., p. 1.
 Ibid., pp. 11-12.
 The exemption provided in section 51(2)(a) was of limited relevance to most labour activity until quite recently. Prior to 1995, the provisions of Part IV of the TPA were limited to corporations; however as trade unions are not corporations, but rather body corporates by virtue of their registration under Industrial Relations Act, Part IV did not apply to their conduct, except in a number of limited circumstances - section 5. The reach of Part IV was extended to include individuals and unincorporated associations by the relevant state Competition Policy Reform Acts, enacted in 1995.
 As the provision was initially drafted it did not apply to employers. However the section was subsequently amended following the decision in Ausfield. In that case the Federated Storemen and Packer’s Union made a demand on Ausfield, an auto part distributor. In order to strengthen its position with Ausfield, the union placed a ban on all Leyland goods handled at the Chullora container depot. Leyland, concerned about the effect of the ban on its business, agreed with the union to stop supplying Ausfield with auto parts – Ausfield Pty Ltd v Leyland Motor Corporation of Australia Ltd (No. 2) (1977) ATPR 40-024 and 40-025.
 Indeed the threshold for invoking the exemption does not appear to be very onerous, given one must only show that a contract, arrangement or understanding was related to the remuneration, conditions of employment, hours of work or conditions of employees.
 NCC Draft Report, Review of Sections 51(2) and 51(3) of the Trade Practices Act, 1999.
 DEWSRB Submission, to NCC Review, dated January 1999.
 See generally McNaughton, op. cit., pp. 117-119; Chin, op. cit., M Costa. "Union Strategy Post the Workplace Relations Act". (1997) 23 Australian Bulletin of Labour (1) 48; V Winley, “Workplace Relations Act: Implications for Business" (1997) 23 Australian Bulletin of Labour (1) 82.
 John Howard cited in Naughton, op. cit., pp. 112-3.
 Section 189(1)(j).
 Section 189 (1C)(c).
 Part XA.
 Part IX.
 Part VID.
 See Costa, op. cit., P Dawkins, "The Tortoise Makes a Move Forward: The Economic Effects of the Workplace Relations Act" (1997) 23 Australian Bulletin of Labour (1) 59; J Sloan. "The Workplace Relations Act 1996: An Overview" 23 Australian Bulletin of Labour (1) 28.