This special issue of the BJIR marks the centenary of the passage through the United Kingdom Parliament of the Trade Boards Act 1909. This statute was not the first enactment of minimum wage legislation in the modern era, for similar laws had been passed in Australia and New Zealand in the 1890s. But among the larger nations this was a first, and it heralded waves of twentieth-century wage regulation around the world, and much associated scholarship. We celebrate this anniversary with a selection of papers from several countries.
The Trade Boards Act, which became law on 20 October 1909, empowered the relevant government ministry of the day, the Board of Trade, to set up a board in any industry in which wage rates were ‘exceptionally low comparedwith that in other employments’. The trade boards resembled joint negotiating bodies, with representatives of employers and workers from the trades concerned, but with additional independent representatives, whose main role was to facilitate agreement between the employer and union sides. To begin with, only four industries were regulated — ready and bespoke tailoring, paper box making, lace finishing and chain making. The powers of the first boards were strictly confined; they could only set minimum hourly rates and equivalent piece rates for their trades. However, there was early evidence of their effectiveness in terms of improving working conditions (Tawney 1914, 1915). In this issue, Jessica Bean and George Boyer review newly collated historical data to examine the boards' impact on poverty. They find that even the limited interventions of the first trade boards materially improved living standards and contributed to poverty reduction through their impact on female wages.
The Whitley Committee, reporting in 1917, recommended an expansion of the trade boards' functions so that they should become ‘a regular machinery for negotiation and decision on certain groups of questions dealt with in other circumstances by collective bargaining’ (Ministry of Reconstruction 1918: para. 11). This change was duly brought about by the Trade Boards Act 1918, and by 1921, there were over 40 boards in place, covering three million workers. Notwithstanding some retrenchment following the report of the Cave Committee in 1922, the powers of the boards were extended during the interwar period to include the regulation of holiday entitlements. The model they embodied was applied to the road haulage and agricultural sectors and, as Boyd Black shows in his paper on a previously under-researched area, to regional contexts such as that of Northern Ireland, where a parallel system to that of the British mainland was put in place notwithstanding strong deflationary pressures and regional pay disparities. The reforming Wages Councils Act 1945 was based on the premise that the state should use its powers not simply to ameliorate the effects of ‘sweating’ (extreme low pay and casualization of employment) but to ‘keep collective bargaining going when economic circumstances tended to destroy it’ (Bayliss 1962: 56). The wages councils, as the trade boards then became, were vested with statutory powers to regulate all aspects of wages, hours and holidays in the affected trades. At this point, approximately one in four of all workers, some 4.5 million, were covered by statutory regulation (Bayliss 1962: 73).
As Sheila Blackburn's paper in this special issue explains, the model contained in the Trade Boards and Wages Councils Acts fell a long way short of establishing a universal legal entitlement to a minimum wage. In Industrial Democracy (first published in 1897), the Webbs had argued for ‘a systematic and comprehensive Labour Code, prescribing the minimum conditions under which the community can afford to allow industry to be carried on; and including not merely definite precautions of sanitation and safety, and maximum hours of toil, but also a minimum of weekly earnings’ (Webb and Webb 1897: 232). As David Metcalf shows in his paper, Fabian writers at this time had a fully worked out conception of the minimum wage which anticipated later debates over levels, enforcement and the relationship between the statutory minimum and the social security system. The 1909 minority report of the Poor Law Commission, largely drafted by Beatrice Webb, saw statutory wage regulation as part of an integrated strategy, along with social insurance, for addressing persistent labour market inefficiencies and related inequalities (Webb and Webb 1909). Several concrete steps were taken towards establishing a national minimum in the early decades of the twentieth century. Most notably, the National Industrial Conference of 1919, set up with the encouragement of the post-war Lloyd George government, put forward detailed proposals for a legally binding 48-hour working week and a universal minimum wage, only for these to falter in the face of employer opposition (Lowe 1978). The solution arrived at by the 1945 Act was a compromise which involved the expansion of the wages councils alongside government encouragement for industry-level, multi-employer bargaining in the voluntary sector. In this way, the state lent its support to the enforcement of basic wage rates and terms and conditions on an industry-by-industry basis, while preserving, as far as possible, the autonomy of the collective bargaining process, a major policy priority at a time of ‘collective laissez-faire’ in industrial relations (Kahn-Freund 1959).
Because the basic terms and conditions set out in sector-level collective agreements could, in the last resort, be made binding on employers by virtue of the arbitration mechanisms first put in place by the wartime Order 1305 and then carried on in postwar legislation culminating in Schedule 11 of the Employment Protection Act 1975, there is a case for saying that the post-1945 system was a universal statutory minimum wage system in all but name. However, the piecemeal approach to regulation, with separate industry-level norms, and the policy preference for voluntarism, shaped both the operation of the system and its outcomes. As the Balfour Committee noted in 1929, ‘the Trade Boards have generally based their rates not on any theoretic minimum standards on life but on the practical conditions of the industry and on a comparison with the rates arrived at by voluntary agreement in other industries for somewhat similar services’ (Committee on Industry and Trade (‘Balfour’) 1929: 90). The wages councils carried on this approach and remained, in essence, an adjunct to the collective bargaining system. There was no attempt to meet what the Webbs had referred to as ‘the paramount necessity of so fixing and gradually raising the National Minimum as progressively to increase the efficiency of the community as a whole’ (Webb and Webb 1897: 779).
Given its structure and function, it is not surprising that the wages councils system struggled to maintain its legitimacy in the postwar years. It was first called into question in the 1960s, when the prevailing view was that the retention of statutory controls was holding back the development of voluntary collective bargaining in the sectors covered by the legislation. In the 1960s and 1970s, 27 wages councils, covering around half a million workers, were abolished. It was later shown that this had not brought about the expected resurgence of voluntary wage determination. The causes of low pay lay not in misguided state intervention, but in industrial structure: ease of entry, intense competition over price, and the availability of casual labour posed enduring obstacles to effective organization on both the worker and employer sides. The authors of this study argued the case for a statutory minimum wage as an alternative to the partial coverage of collective bargaining (Craig et al. 1982). But by the time their research was published, the policy pendulum was already moving in the direction of labour market deregulation. The Wages Act 1986 removed the powers of the wages councils to set more than basic time and piece rates, in the process eliminating all statutory paid holiday entitlements. Complete abolition of the remaining 26 councils followed in 1993. The wages council system was far from irrelevant at this point; prior to abolition, it covered 2.5 million workers, or around 10 per cent of the active labour force (Deakin and Wilkinson 2005: 271). The change was not primarily the result of employer pressure; smaller firms, in particular, relied on wages council rates to standardize pay arrangements and minimize bargaining costs (Bryson 1989). Employer lobbying was sufficiently strong in agriculture to lead to the retention of statutory wage fixing in that sector.
This time of retreat on regulation coincided, paradoxically, with renewal in the economic analysis of the effects of minimum wages. Against the then-prevailing view that raising wages induced employers to reduce employment, the extent of loss depending only on the elasticity of the labour demand curve, new non-competitive models of the labour market, including those premised on asymmetric information and dynamic monopsony, were laying a claim to represent the reality of many low-skilled labour markets better than the textbook perfect-competition model. Dynamic monopsony theory introduced a substantial element of ambiguity, even suggesting that moderate rises in minimum wages would increase employment because of their positive effects on labour supply flows (see Manning 2003). Simultaneously, advances in empirical research techniques and the availability of new sources of data revealed instances of minimum wage increases which induced only very modest changes in employment, either positive or negative (Card and Krueger 1995; Machin and Manning 1996). One technique in growing use was meta-analysis. In this issue, Hristos Doucouliagis and T. D. Stanley update, correct and corroborate the findings of earlier work demonstrating that minimum wage studies reporting employment reductions were contaminated by publication selection bias. After controlling for selection, they find little evidence remaining of any negative association between minimum wages and employment. Two other papers add to this literature in focusing on specific groups: younger Finnish workers in the case of Petri Böckerman and Roope Uusitalo, and older Canadian workers in the case of Tony Fang and Morley Gunderson. Both these studies fail to find any significant negative impacts on employment, and in the Canadian case there is some evidence of a modest positive effect.
In the early 1990s, research on British wages councils was finding no beneficial employment increases from the reductions in wages that had taken place throughout the 1980s (Dickens et al. 1993, 1995), and was already questioning the case for abolition. It was not long before new legislation was in place. The National Minimum Wage Act 1998 enacted a model of wage regulation which, on the face of it, more closely resembled the universal minimum which the Webbs had argued for than the wages councils system did. There was now a statutory, national minimum wage (‘NMW’), binding on all employers regardless of the sector they were based in. The statutory minimum was expressed as a single adult hourly time rate, initially set at £3.60, with lower rates for younger workers and apprentices. Detailed statutory provisions were put in place in an attempt to ensure the application of the basic time rate to piece work, salaried work and other working time arrangements, to deal with the question of deductions for living expenses and other allowances, and to provide for enforcement (Deakin and Morris 2005: 285–92; Simpson 1999a,b). In each year since its inception, the minimum rate has been raised, mostly in line with or above retail inflation. The legal power to raise the minimum rate vests in the Secretary of State for Business Enterprise and Regulatory Reform (the successor, in this respect, to the President of the Board of Trade) who may consult the Low Pay Commission, a statutory body with tripartite (i.e., employer, union and independent) representation, when doing so, but is not required to.
Appearances can be misleading. The 1998 Act, while formally resembling a universal minimum, owes much to the wages council system and even more to the original trade boards model of 1909. The tripartite structure of the Low Pay Commission is in a direct line of descent from the arrangements put in place for the trade boards. As in 1909, the 1998 Act and the related National Minimum Wage Regulations 1999 contain no power to set the minimum wage at a level which reflects living costs. In contrast to the long-standing position with regard to social security benefits in the UK, there is no statutory mechanism for automatically uprating the minimum wage with prices. There is an even more marked contrast here with statutory minimum wage regulation in some other European countries, most notably the French model of the ‘minimum growth wage’ (the salaire minimum interprofessionel de croissance or ‘SMIC’). The law governing the SMIC, which has been in force since 1970 (following an earlier law on minimum wages of 1950), links the minimum rate to price inflation, and also makes provision for it to be raised each year by at least half the increase in the value of the purchasing power of the average wage (see Pélissier et al. 2008: 958–63). The French model was influential in British debates over low pay in the late 1980s and early 1990s, in particular in the work of the Low Pay Forum, a co-ordinating group whose members included union representatives, parliamentarians, academic researchers and the Low Pay Unit. The Low Pay Forum's programme for implementing an NMW, published in 1988, included a proposal for an uprating mechanism which was designed to maintain the value of the minimum rate in line with general movements in wages (McNeil and Pond 1988). It was also notable for addressing the relationship of the statutory minimum wage to other, complementary labour market institutions, including sectoral collective bargaining, the tax and social security system, equal treatment legislation, and active employment policy. The Forum's proposals helped to shape the Labour Party's manifesto commitment to the introduction of an NMW before the General Election of 1992, which Labour lost. The recommendation for an automatic uprating mechanism did not survive into the proposals which the party drew up while in opposition in the mid-1990s and which were implemented, after it returned to power, in the Act of 1998.
Under the regime of the 1998 Act, the absence of an automatic uprating mechanism has been remedied, to some degree, by the practice of raising the rate in line with, or in some years above, the consumer price index. However, only a single minimum rate (with variants for younger workers) can be set, a throwback to the limited provisions of the 1909 Act, in contrast to the powers to fix statutory wages above the basic floor and to regulate terms and conditions of employment more generally which were provided for in the Act of 1945. As before, close regard is paid in the wage-fixing (or, strictly speaking, wage-recommending) process to the need not to set the minimum at a rate which departs too far from industry practice. In this respect, there is a critical difference between the current regime and its predecessors. Even under conditions of wage-cutting and downwards pressure on terms and conditions in the interwar period, it was the rates set by voluntary collective bargaining to which the trade boards had regard, as the Balfour Committee recognized. Today, following the dismantling of the structure of sector-level bargaining from the 1980s onwards (on which see Deakin and Wilkinson 2005: 268–9), there are few relevant industry-level rates against which the legal minimum can be compared.
The principal constraint on raising the statutory minimum is no longer the policy goal of preserving the voluntary collective bargaining system, but the perceived need to minimize what are seen as potentially negative economic impacts, in the form of inflation and unemployment, of this form of statutory intervention in the employment relationship. In this respect, the research generated by the Low Pay Commission, and the use made of it, has been one of the better examples of evidence-based policy making: it has supported the attempt to neutralize and rationalize the process of setting the rate of the minimum wage. In this issue, William Brown's account, from the ‘inside’, of the deliberations of the Low Pay Commission makes the case for it having made a positive contribution to the practice of social partnership in Britain, as it balanced competing interests, sustained its authority and yet kept itself autonomous from government. Independent research appears to have been prominent in this process.
The consensus view which has emerged from a wide range of empirical economic studies is that the introduction of the NMW has not had the negative effects which orthodox economic theory suggests it should have had (Metcalf 2008). Rather, the evidence is mostly in line with the average estimates from multiple studies reported in meta-analyses. However, this type of finding does not necessarily shake the basic position of mainstream neoclassical economics; it is possible to interpret the recent UK experience as showing that negative impacts are possible, but that their magnitude may be small, or localized to particular sectors of the economy, depending on context. Because the context for the NMW has, until recently, been remarkably favourable — a decade of sustained, non-inflationary growth, coupled with rising employment — its experience, while reassuring for the advocates of statutory labour market regulation, might be regarded as atypical. Proponents of a laissez-faire approach cannot be expected to regard it as convincing evidence for the positive economic effects of the minimum wage.
And yet, the arguments on this issue should ultimately transcend the question of whether minimum wages benefit the low-paid, and if so, disadvantage the unemployed. Bruce Kaufman's paper in this issue suggests that, in the debate over the economic effects of the minimum wage, the wrong question is being addressed. He argues that some disemployment effects are to be expected from the implementation of a statutory minimum, as less efficient forms of production are priced out of the market. However, this is an essential part of the social cost argument in favour of the national minimum which was advanced by the Webbs. The minimum wage removes the artificial subsidy which low pay provides to inefficient firms, and which in an ‘unregulated’ system is borne by other firms and by the community at large. In this way, it removes an externality and realigns the industrial structure with the wider interests of society. This argument has not featured much in the recent debate over the impact of the NMW, but it has long formed part of the approach to wage regulation on the continent of Europe, as Colm McLaughlin's comparison of the Irish and Danish cases makes clear. An implication of McLaughlin's analysis is that minimum wage arrangements are, to a certain extent, endogenous to the economic and social context at national level, as well as being path-dependent in the sense of being shaped by prior political choices and inherited institutional structures. The extent to which the institutional form taken by the minimum wage in a given country is a reflection of wider social and political forces is also highlighted by Rüya Koçer and Jelle Visser's paper. Comparing the United States and Turkey, they show that the low level of the US minimum wage is linked to the very small percentage of the workforce who receive it (2–3 per cent) and to their lack of political influence. In Turkey, by contrast, where the minimum wage serves as a benchmark both for non-unionized workers in the formal economy and for non-registered workers in the informal economy (for whom it operates as a maximum), the large numbers directly affected by the minimum wage fixing decision have led to a situation in which considerations of legitimacy play a prominent role in the setting of the minimum rate.
One of the beneficial side effects of the intense interest in the effects of minimum wage regulation over the past decade, in Britain as elsewhere, has been the growing sophistication of social science research on this subject. Theoretical models are both more rigorous and more realistic than they used to be, new data sources have been used in an imaginative way and comparative analysis is throwing light on the causes and consequences of cross-national diversity in minimum wage arrangements. The general trend of this research is to stress the potentially positive economic impacts of minimum wage regulation in terms of the reduction of social costs and the promotion of partnership at work. Public policy, at least in Britain (but not only there), has yet to take this body of work fully on board. A century on from the first Trade Boards Act, perhaps the time has come to look again at the Fabian argument for the ‘public organisation of the labour market’ (Webb and Webb 1909).