This article explores the UK's National “Living” Wage framework from a theoretical, and historical, perspective. It analyses how economic theories of the wage are reconstructed in juridical terms in the context of policy and legislation. By comparing two approaches to economic analysis, neoclassical economics, and classical political economy, it will show that how the labour market is conceived influences our conception of the minimum wage and of the function of labour law in a market economy. These insights, it will be shown, can be used to shed new light on how UK minimum wage law has developed, so that we can better understand UK “living wage” legislation today.
The article begins by distinguishing two competing conceptions of the minimum wage in economic theory.1 These are, first, the neoclassical conception of the minimum wage as a market-correcting device, designed to restore the market to “equilibrium” (a “market wage”); and second, the conception of the minimum wage we find in classical political economy, a “natural” or “social wage” that reflects workers’ customary living costs. These two approaches reflect two of the “wage's” different (inter-dependent) functions: as the price of a commodity, labour power, which is exchanged in the labour market and, as with any other “price”, functions to coordinate decision-making; and second, as the cost of reproducing that commodity, a process which is not confined to the market but takes place in society more generally: the function of social reproduction. The article will show that the benefit of the classical political economy approach lies in recognising that these two functions are both conceptually distinct, and frequently in conflict. The price the market assigns to the labour commodity is not always, and not necessarily, that which is required to cover its costs of (re)production. By focussing exclusively on the exchange-process, however, this is an observation that neoclassical economics tends to overlook. This has led neoclassical economics to adopt a narrow conception of the minimum wage as a mechanism that plays a market-correcting role. By leaving it to the market to decide how resources should be allocated in society, this perspective implicitly marginalises distributional and fairness concerns from the scope of the minimum wage.
Engaging with the concept of the “natural” or “social” wage as developed in the works of the classical political economists, however, opens the door to a much broader, constitutive role for labour law, and a broader conception of the minimum wage. Here, the minimum wage can be embraced as a mechanism capable of expressing fairness norms, channelling the distributional conflicts that are inherent in a capitalist (or market society) so as to secure term social and economic sustainability.2
Having explored the minimum wage through the lens of economic theory, the article goes on to explore how legal conceptions of the minimum wage have been expressed in statute and case law concerning minimum wages over time. It shows that the economic categories of the “market wage” and “social wage” are not just theoretical constructs; they have historically found (approximate) expression through the legal concepts of the “wage”, a payment for work that is linked to time spent working, and, building on the older concept of the “salary”, the concept of “remuneration”, a contractual payment for employment over time. Minimum wage legislation that has used the juridical concept of the “wage”, such as the Trade Board Acts and the National Minimum Wage Act, has thus tended to be linked with labour market policies concerned primarily with ensuring fair competition and efficient resource allocation in the market. Minimum wage legislation that draws on the concepts of “remuneration”, such as the Wages Councils Act 1945, by contrast, has tended to go much further, linking workers’ income with communal and political expectations of workers’ living standards, and shifting some of the social costs of employment onto employers.
These observations highlight that the choice of concept used in legal drafting is important when it comes to shaping how a particular policy is interpreted and applied in practice. Legal concepts are not empty shells that it is for policy-makers to fill with legal content; history and the structure of legal doctrine both play an important role in shaping their meaning and scope. If attention is not paid to the meanings embedded in concepts, such as the wage, therefore, policies may have unforeseen consequences, and/or fail to achieve their intended effects.3
The insights that will be developed in the course of this article tell us much about the conception of the minimum wage embedded in the national “living” wage framework which was introduced into UK legislation from 2016. As we shall see, the National Minimum Wage Act and Regulations create a link between the right to be paid and the obligation to work. As a result, their shared juridical structure embodies a market, rather than a social wage, conception of the minimum wage: that is, a wage tied to private marginal costs, rather than to worker's costs of living. This market-based conception of the wage, in turn, has underpinned a major and expanding role for the tax credit system when it comes to providing for workers’ subsistence.
The article will go on to show that, despite appearances, the introduction of the National Living Wage rate and associated changes to the tax-credit regime in 2016 did not usher in a new “social wage” model of minimum wages. This is because, while the National Living Wage altered the arithmetical relationship between the minimum wage and tax-credit regime (or, as it is in the process of becoming, Universal Credit), it did little to change the conceptual foundations of the National Minimum Wage Act and Regulations, which together continue to tie the right to be paid to the provision of “work”. The National Living Wage, as currently operationalised, is ill-suited when it comes to providing the sort of “social” wage that, a fair, efficient and sustainable, labour market requires.
II. Economic Theories of the Labour Market
A. Neoclassical Economics
The modern discipline of economics is principally concerned with the question of how to secure an optimal, or efficient, allocation of resources in conditions of scarcity. It is thus a study, first and foremost, of the sphere of exchange, with questions of efficiency conceived in terms of how to secure an optimal usage of resources in order to maximise individual (and thereby aggregate) welfare. This is deemed to require that prices reflect the private (marginal) costs of production, so that “supply” equals “demand” – a task which, for neoclassical economics, is achieved through a decentralised pricing mechanism in the context of a free, perfectly competitive market.4
Neoclassical economics extends its model of the competitive market to the study of labour markets: treating labour like any other factor of production.5 In this context, competition between workers and firms will ensure that wages will cover the workers’ costs of living: if wages are too low, workers will refuse to work, there will be a shortage of labour relative to demand, and workers will be able to bid up wages until supply and demand are in equilibrium.6 If wages are too high, by contrast, employers will be able to bid wages back down due to increased worker competition. Left to its own devices, the market dynamic, driven by the rational utility-maximising decisions of economic agents, will keep the wage in the appropriate market-clearing range so that the production costs of firms, and workers, are fully covered: prices for labour of comparable productivity will be more or less equal throughout the market, and so will be equivalent to private marginal costs of production.
In this account, wages are governed entirely by the relationship between supply and demand. The supply of labour is thus entirely governed by markets (rather than institutions) and the same goes for its allocation: prices dictate how labour will be allocated, and used, throughout society, by telling workers and firms how much it is “worth” in terms of utility. Workers decide to produce more or less labour (power) depending on the relative benefit of an additional hour's work, as expressed in the wage (in light of prevailing prices of consumption goods), as compared with the “opportunity” cost of lost leisure time. Firms will demand more or less labour depending on the relative value of the worker's output (his marginal product), given prevailing wage rates, and the likely returns from alternative methods of production (such as machinery). In the absence of legal intervention, the prevailing wage rate acts as a reliable guide as to what amounts to a “fair bargain”, so each party can be confident that they have received something of equal value to that which they provide in exchange.7
The “ideal” neoclassical market exists in a world of zero transaction costs and perfect information.8 In this context, the introduction of a minimum wage would be detrimental to efficiency because it interferes with the market's process of self-adjustment, leading to a mis-allocation of resources and, ultimately, unemployment.9 In a world of positive transaction costs, however, disequilibria can, and will, result. This is particularly so given the prevalence of employer monopsony: in a number of industries, a single, or small number, of employers will dominate, exercising such a degree of market power as to be able to offer wages far lower than they otherwise could in highly competitive conditions.10 In this situation, even the neoclassical school can accommodate a minimum wage, provided that it is carefully designed to target specific market failures, and so is set no higher than the equilibrium rate (i.e it is within the market-clearing wage).11 Tailored in this way, minimum wages merely restore the market to equilibrium so that supply can once again equal demand.12
Neoclassical economics is not entirely averse to mandatory minimum wages, therefore, provided they can be precisely tailored to target identified market failures. This leads, however, to a very narrow conception of the minimum wage as a “pseudo” market wage, set at the “competitive” rate, and so set independently from considerations about worker's living costs. It is also, by implication, entirely blind to fairness norms prevailing in society, and the customs, and expectations, of the working classes. Because markets are spontaneous orders that “exist” in a state of static equilibrium, a minimum wage that goes beyond a narrow market-correcting role, however, that attempts to raise wage rates or to express considerations of fairness, will, by interfering with the market's (inherently efficient) allocation of resources, be harmful to society as a whole.
B. Classical Political Economy and the “Natural” Wage
The Classical Political Economists took an entirely different perspective on the nature, and origins, of markets.13 For scholars such as Smith, Ricardo, and Marx, economics was not the study of the market process, but of the process of social reproduction; production, consumption and exchange all had to be taken into account.14 Markets were not spontaneous orders but instituted processes, profoundly shaped by politics, custom and the law.15 Considerations about efficiency were not assessed in terms of how to maximise the (aggregate) utility of economically rational agents, therefore, but with the broader question about how to secure the long-term sustainability of the market (or “capitalist”) system.16 This required more than simply aligning prices with private marginal cost; it required the reproduction, and increase, in the existing stock of productive resources so as to fuel future production, and consumption, and raise the living standards of the general population.
Because they studied markets in their social context, the classical economists did not see wages as the product of supply and demand; bargaining inequalities were key.17 Combinations among workers were seen as important institutional elements acting, in conjunction with the forces of supply and demand, to determine the relative strength of the parties.18 This relative bargaining strength was what determined the normal range around which competition would act.19 By reason of worker's wage dependence and the superior economic strength of employers, however, in the absence of comprehensive collective bargaining and limits on employer combinations, it was assumed that market wages would inevitably diverge from the “natural” or normative rate required for social reproduction.20
This recognition, of the embedded nature of markets, led the classical economists to distinguish between the actual costs to society of producing commodities, their “natural” or social price, and the value of those commodities as expressed in market-prices. In the labour market, this was reflected in a distinction between the “natural” and the “market” wage21:
The natural price of labour is that price which is necessary to enable the labourers, one with another, to subsist and to perpetuate their race, without either increase or diminution …. The market price for labour is the price, which is really paid for it, from the natural operation of the proportion of supply to the demand .… It is not to be understood that the natural price of labour, even estimated in food and necessaries, is absolutely fixed and constant …. It essentially depends on the habits and customs of the people.22
For the Classical Economists, this “natural wage” was the normative standard around which market wages gravitated, a reflection of prevailing fairness norms, and the outcome of past conflict.23 Because it was established outside the market, determined by customs, politics and institutions, it was independent from the short-run conditions of supply and demand. Thus, while the market wage might tend, in practice, towards this “natural”, or “customary” level, deviations were inevitable. This was due not to “interferences” in the market's spontaneous processes, but to the unique nature of the labour commodity and its process of (re)production.24
The Classical economists recognised that labour is different from other commodities because it cannot be separated from its owner.25 For this reason, labour supply is not dictated by market forces, but by social institutions, such as the family; as well as the customs and habits of the working classes.26 The question to be addressed, therefore, was whether, and if so how, these processes might come to be reflected in the wage rate so as to secure social reproduction in practice – given that, in a capitalist system, employers have little incentive to pay “natural” or “social wages”, and workers lack the bargaining power to demand them.27
The neoclassical model largely ignores these questions because it starts from the premise that labour (power) already exists as a commodity that can be sold by free economic agents in a competitive market. Once the worker's wage dependence is taken into account, however, the bargaining process appears somewhat different. If workers have no choice but to work for someone, “rational” utility-maximising behaviour will not, in itself, ensure that information about worker's “production” costs are embedded in the wage rate. We cannot assume that workers will refuse a wage that falls short of the costs to them of working; the opportunity cost of lost leisure time. In conditions of wage dependence, additional “leisure” time is rarely an option: workers face a choice between accepting the wage offered, seeking out alternative employment or, in modern times, relying on the family, or the state, to support them. At best, therefore, the wage rate will reflect the best wage offered by the firm's competitors, or, in modern conditions, the prevailing rate of social security payments (the worker's “best” available alternative); actual living costs will not necessarily be incorporated into the wage rate.
It may be, as Smith argues, that if “masters … always listen[ed] to the dictates of reason and humanity” they would willingly pay subsistence wages – otherwise, they risk a depreciation in the stock of resources available for productive use.28 But not only does this overlook that their ability to pay such a wage itself requires some institutional mechanism by which to determine this “social” wage rate, but also, market competition is such that no single employer has an incentive to pay such wages unless other employers are doing the same.29
The problem, as Smith agued, is that, in a capitalist labour market, workers do not sell themselves outright, they hire themselves to an employer for a particular period of time. This means that employers can make use of their labour without having to worry about the long-term effects on their physical, and psychological health – this is what distinguishes wage-labour from slavery.30 In the case of slave labour, slaves are an owned asset much like machines; employers stand to lose, therefore, if they do not cover the costs involved both in the day-to-day use of the slave, and the long-term costs involved in securing him a long, and healthy, existence.31 In the case of wage-labour, while employers have an incentive to provide workers with enough money to fuel them throughout the working day, they have no incentive to cover the “fixed costs” of maintenance and repair32 – to preserve the value of their human capital over time; workers can be readily discharged or replaced.
While neoclassical economics is largely blind to this social cost gap, (a consequence of their supply and demand model of markets), the Classical Economists were not.33 It is reflected, for example, in Marx's distinction between labour and labour power, and his analysis of “surplus labour”, and in the arguments of Smith and Ricardo that the gap between the natural and market wage would have to be closed if existing populations levels, and the existing rate of profit, were to be maintained.34 The natural or “social” wage was thus seen in classical political economy as a de facto wage floor below which the market could not be allowed to fall: subordinating the process of capital accumulation had to be subordinated to the requirements of social reproduction.35
Smith and Marx both expressly defended as system of high wages on not just moral but also efficiency grounds: legislation mandating “social” wages was in the best interests of society and conducive to economic growth.36 For each, law, and institutions, did more than “correct” the market: they were integral to its functioning. As Boucoyannis argues37:
“Price limits are misguided only where competition can respond to change and local conditions more effectively” … the38 key criterion between good and bad intervention is securing the most productive use of capital – and not burdening the poor. Regulation of the right kind is thus necessary in [Smith's] “system of liberty” … the institution of the market has to be “made” – only the practice of barter, as well as collusion and deceit, emerges spontaneously.39
In classical political economy, labour markets were not spontaneous orders in which legal regulation “intervened”. Rather, they were social institutions, with law playing an integral part in the constitution, and reproduction, of the labour commodity and in the conditioning of wages. Contrary to the assumptions of the neoclassical school, therefore, labour power was not, and could not be, treated, institutionally, like any other commodity. Risk shifting mechanisms, such as sick pay and pensions, legislative support for collective bargaining, and/or a minimum wages tied to living costs, can, from this perspective, be seen as part of the broader institutional framework that makes the commodification of labour both possible and sustainable.
C. Rethinking the Minimum Wage
Because the classical political economists saw labour markets as social institutions, they were able to embrace the complex factors influencing the supply, allocation, and use of labour power in the economy. Labour supply was not governed by rational decision-making, but by social norms, customs, and institutions, including labour law, social security systems, and the family. Decisions about who to hire, and at what price, moreover, were institutionally conditioned, influenced by social norms and expectations and the balance of power between labour and capital. Labour power was not a “natural” commodity, but something that was treated as such only in certain historically specific conditions.
By adopting the market-based supply and demand model, by contrast the neoclassical economists concealed labour's costs of reproduction, and the institutions that are required to support it. By taking the existence of markets as given, it can admit only a limited conception of labour law as a mechanism for correcting market failures. Only a very tightly circumscribed minimum wage, tied to the market-clearing rate, can be permitted in the name of efficiency, which, in the neoclassical frame, is conducive to “fairness”. By confining employers to paying the private costs of labour, however, this leaves the social cost problem unaddressed, shifting these costs onto the family and the state. In modern times, this problem has tended to be addressed through wage supplements and tax credits – funded by the tax payer. By leaving employers free to exploit social resources at a fraction of the cost, however, this simply provides a subsidy to inefficient firms which would otherwise be forced to improve their production methods to compete. By interfering with the information embedded in wages, it risks over-production (due to artificially low production costs): wasting resources that could be better used elsewhere.
If the labour market is understood as an instituted process rather than a spontaneous order, however, then regulation is necessary to constitute, and not simply to adjust, or “correct” it.40 From this perspective, labour law is best seen as something that helps constitute the labour commodity, and regulates the resulting social and economic relations.41 The specific form that regulation takes depends on the social, political, legal and cultural context as well as the balance of power between labour and capital, and between different segments of the labour market.42 But how the law constructs the employment relationship inevitably influences the balance of power between capital and labour, and in this way, the wage rate. Even the “market” wage cannot be separated from the institutional processes through which labour power is produced (and reproduced), therefore, and which, in turn, constitute, and regulate, the labour market.
We must take seriously, therefore, the role that custom, politics and institutions play in conditioning the supply of labour (and thus, the costs of production), and its price. Because the production of labour power takes place in part outside the market, the supply of labour cannot be governed entirely by a mechanical relationship between prices and quantities in exchange: customary standards of living, hours, inclination to work, family structures and power relationships and the broader legal and institutional environment, all have to be taken into account.43 These external considerations cannot be expected to be reflected in the wage rate “automatically”: in the real world, characterised by imperfect information, wage dependence and an inequality of bargaining power, how much workers will accept as wages will not be the same as how much they need to be provided with to cover their costs of reproduction. Thus, what is needed is a mechanism, or set of mechanisms, capable or imposing on employers an obligation to pay a wage that reflects workers’ expectations as to their standards of living, transcending distributional conflicts, reflecting fairness norms prevailing in society – a function which has, historically, been somewhat imperfectly performed through labour law and the contract of employment.44
III. The Evolution of the Minimum Wage as a Legal Concept
This section explores how the economic categories, or functions, of the “market” and “social” wage are reconstructed in juridical terms in the context of UK minimum wage law. Rather than implying a direct correspondence between legal and economic categories, what this section suggests is that, because these “functions” reflect real problems inherent in capitalist labour markets, and because legal concepts are forged through attempts to operationalise the law in light of such problems, we can expect some correspondence between them.45 By uncovering the assumption embedded in the concepts through which minimum wage law is given effect, therefore, we can further our understanding of how well English minimum wage supports the goal of dynamic, long-run sustainability.
A. Origins of Minimum Wage Legislation
The starting point for a modern history of English minimum wage legislation is the Trade Boards Act 1909.46 This Act emerged as a response to the problem of “sweated labour” but fell short of the statutory minimum “social” wage advocated by social reformers such as the Webbs.47 Rather, the legislation simply placed on a statutory footing joint regulation in “sweated” trades where collective organisation had either failed or broken down, and where wages were “exceptionally low as compared with that in other employments”.48 The Act empowered these Boards to make provision for minimum rates to be set by reference to existing practices in the organised sectors.
From the perspective of the Act, the right to the minimum wage was no more than a right to be paid the exchange value of a commodity. The purpose of the Act was to correct a perceived market distortion, to prevent employers from gaining a competitive advantage by paying less than the true market price.49 For this reason, the Board's powers were confined to setting minimum rates for labour by reference to prevailing practices. They thus had no power to set minimum terms and conditions of employment apart from wages, to provide workers with a guaranteed minimum amount of work or minimum income from employment, nor to take account of the costs of living when setting the rates.50 The Boards were simply intended to “give statutory sanction to principles and methods adopted in numberless cases by voluntary agreement”.51
This idea of the wage, as a price, stands in sharp contrast with the concept of the salary that was used to refer to the payment made to professional and public office holders in exchange for their service over time. Salaries differed from wages in that they were fixed and unconditional payments, set at a level that reflected the dignity, and duties, of the office or profession of the person to whom it was paid.52 They were individualised payments, therefore, and did not vary depending upon how much work was performed. They would also often be supplemented by a number of additional monetary or non-monetary benefits which, along with the salary, made up their overall “remuneration” for employment,53 the sum that could be recovered in the event that the employee was wrongfully dismissed.54
Prior to the late nineteenth century, the status, and working conditions, of manual labourers and professional employees hired under a relational contract of employment could not have been more different. There was no suggestion, therefore, that the type of payment paid to a higher-status worker might be a “wage”, nor that a “wage” might be paid to anyone other than those who, absent such wages, “would be likely to be deprived of their daily means of subsistence”.55
This position began to change in the late nineteenth /early twentieth century. The spread of collective bargaining, social welfare legislation, and changes in company law promoting vertical integration, had implications for the employment conditions of all workers, and this had the effect of blurring the distinction between wage-workers and salaried employees. One of the most significant developments during this period was the introduction of Workmen's Compensation Legislation in 1897. The purpose of the Act was to shift onto the employer some of the risks associated with industrial employment. The Act made provision for compensation to be paid by employers to workers injured at work, and this would be calculated by reference to “earnings in the employment of the same employer”.56 The Act only applied to manual workers, but it presupposed a relationship in which there were mutual expectations of ongoing employment – it sought to compensate workers whose employment had been terminated prematurely or interrupted due to industrial injury.
The relational dimension implicit in the Act had important implications when it came to how the term “earnings” was construed. In contrast with “wages”, in the context of the Act “earnings” had to be understood “not in the sense in which economical writers use it, but in a popular sense … the full sum for which the man is engaged to work”.57 This was not the same as the price of personal labour, the wage, but corresponded with the total remuneration that was payable to the worker in connection with his employment, anything that the worker had a legitimate expectation to receive, whether from his employer or otherwise.58 Implicitly, therefore, employers of even manual workers were now seen to owe obligations that extended beyond the obligation to pay wages, and this included the provision of benefits and allowances that were not directly linked with how much work was performed.
The emerging distinction between the concepts of the wage and remuneration became increasingly significant throughout the course of the nineteenth century, as the courts elaborated the latter concept in light of principles associated with the concept of the salary. Whereas the term wage continued to be associated with the “price” at which labour was sold on the market, the term remuneration was gradually being used to refer to the payment that employers were obliged to pay as a matter of law in exchange for their worker's service. By implication, if Parliament used the term wage, this would severely circumscribe the scope of the statute's protection. By using the term remuneration, therefore, the effect was to provide for a form of legally constituted “social wage” informed not by the intrinsic value, or marginal utility, of a commodity, but considerations about the (customary) costs of living.59 This was particularly so given the tendency, throughout the twentieth century, for even manual workers to be paid a number of fringe-benefits and other benefits not directly proportional to time working; payments which would, in other words, have fallen outside the scope of the narrow concept of the “wage”.
In Roberts v Hopwood, the use of the term “wage” in the Metropolitan Management Act 1855 had the opposite effect, curtailing the scope of the employer's payment obligations by actively preventing it from paying more than the value of the labour provided. Section 65 of the MMA empowered the local Council to “employ such servants as may be necessary, and to allow such servants … such wages as [the Council] may think fit”. In light of this provision, the Council purported to set a flat-rate “wage” for all of its council workers, which meant paying male and female servants the same wage regardless of output. This was not, according to the court, a wage, in the sense of “the reasonable pecuniary equivalent of the services rendered”.60 The rates “were far in excess of those [sums] necessary to obtain the services required and to maintain a high standard of efficiency”, and, set at a flat rate, could not be said to be measuring the “value” of the labour provided. In many cases, therefore, they would manifestly exceed the “market price”.61
In light of this, it is highly significant that the Trade Board Act used the term “wage” rather than remuneration to refer to the minimum rates. In France v James Coombes & Son (1929), for example, the claimant's right was construed as a right to be paid the minimum rate for time spent producing the “good” that was characteristic of the trade to which the Wage Order applied. He could not, therefore, claim minimum “wages” in respect of time performing quasi-managerial duties as part of his employment.62
In his dissenting judgment, Lord Blanesborough advocated reading the Act purposively, interpreting the “wage” as “remuneration” so as to ensure that “the worker shall receive at least the minimum rate of remuneration for the work actually done, or for the time spent in statutory employment”.63 Lord Blanesborough argued that this formulation better reflected the policy behind the Act. In circumstances where the employer's rights were so extensive and exclusive as to prevent the worker from accessing his subsistence elsewhere, the employer had to pay for the full benefit he received: his worker's ongoing service:
during the periods when the employer is not bound either to employ or pay him he must be left at liberty either to obtain his minimum wage from another employer, or to exercise his skill for his own benefit. In no other way can the minimum or subsistence wage which for workers of his trade the Acts essay to provide be found for him …. [This required that] for the whole time of his service there [be] payable to the worker a wage at the minimum time-rate.64
Throughout the early 1920s and 1930s, Parliament remained hostile to the idea of legislating for a right to a minimum level of contractual remuneration. In 1929, for example, it rejected a proposal for holidays with pay on the basis that statutory right to paid holidays would distort the price of labour, forcing employers to lower basic wage rates.65 By the late 1930s, however, political opinion was changing and in 1938, Parliament introduced what might be seen as the first minimum “remuneration” legislation, the Road Haulages Act 1938.66 Rather than regulating competition, the expressed purpose of the Act was to guarantee workers received from their employers something approximating a “living wage”. In order to achieve this, section 10 of the Act conferred on the Wage Boards a power not just to set piece and time rates, but also overtime rates, and holidays with pay.
Experience during the second world war further cemented the premise that workers ought to be provided with a guaranteed minimum income from employment – irrespective of how much “work” they provided. These premises were reflected in the Essential Work Order 1940, enacted with a view to guaranteeing continuous production during the war period. The idea behind the Order was that “The Minister will be able to prescribe the terms of remuneration, the hours of labour, and conditions of service. Remuneration will be on the basis of the remuneration for the job”.67
During the war, therefore, all workers received remuneration for their contribution to the war-effort; their ongoing service to the country.68 They did not receive wages in the traditional sense, because “wages and profits were under the government's control” – their pay was determined out-with the context of the market.69
The war had also been significant, however, for eroding customary wage rates and rigid controls over training, further undermining status distinctions between wage-workers and salaried employees. The high-point of this process came in the mid-1940s with the passing of the Wages Councils Act 1945 and the Catering Wages Act 1943. The Wages Councils Act provided:
for the regulation of the remuneration and conditions of employment of workers in certain circumstances, and for the establishment of Wages Councils with the power to set not just wages, but holidays with pay, overtime rates, as well as powers to prescribe certain minimum terms and conditions of employment. Rather than targeting only the most extreme instances of low wages, Wages Councils were to be established in industries in which existing machinery for setting remuneration was inadequate, was likely to become inadequate or if a reasonable standard of remuneration was unlikely to be maintained.70
In contrast to the Trade Board legislation, the Act was expressly concerned with what the worker received from employment, rather than what the employer paid for labour in the market. The Wages Councils Act expressly defined remuneration as “the amount obtained or to be obtained in cash by the worker from his employer … in connection with his employment”.71 By making employment rather than labour the reference point for the payment of the rate, moreover, the Act firmly embedded the link, emerging in the common law, between the right to a reasonable remuneration and the existence of a particular type of relationship characterised by faithful service and stable employment.
Like the Road Haulage Act the provisions of a Wages Regulation Order was enforceable through the contract, such that “if a contract between a worker and employer to whom an Order applies provides for the payment of less remuneration than the statutory minimum, it shall have effect as if it provides for the payment of the statutory minimum”.72 Provided there was a contract, therefore, of any type, whether of employment or for labour, the worker would be entitled to be paid the “reasonable” remuneration established by the Councils – what both sides of industry regarded as fair or adequate, taking into account the costs of living.
B. Towards the National Minimum Wage
Support for the Wages Councils system remained strong until the end of the 1970s.73 Employers supported the system because it was expedient and helped them to avoid complex local level wage-bargaining, and provided important protection against unfair competition – the driving down of workers’ wages.74 Even the Trade Unions, hostile at first to the threat the Councils posed to voluntary bargaining had, by the 1970s, started to support the system, arguing for its extension alongside a new statutory minimum wage.75
The 1979 Conservative Government had an entirely different outlook, however, when it came to socio-economic policy. Inflation, coupled with rising unemployment, had helped foster support for the belief that market forces were the most efficient mechanism for allocating resources.76 Both the regulation of terms and conditions of employment and collective bargaining had distortive, and inflationary effects and the Wages Council system embodied both. Imposing on the parties terms to which they had not expressly agreed, the Councils set wages above what was seen as the “natural” market-rate.77 From 1985, therefore, the Government launched an attack on the Wages Councils that led to their gradual scaling back and eventual abolition in 1993.
During the same period, the Government introduced a new wage supplement, the Family Credit, an extension of the Family Income Supplement introduced as a temporary measure in 1971. The purpose of the credit was primarily to subsidise employers, to encourage them to hire workers at rates they could “afford”78: “The real policy underlying [the credit] is to try to force down wage levels by paying family credit through the employer, thereby encouraging employers, in addition to the other methods that the Government are using, to force down the wages of the low paid.”79
The 1970s and 80s thus reflected a return to the pre-1940s belief in the existence of an objective, market-clearing wage which ought not be artificially raised to take into account considerations about the costs of living. The result was a tax-credit system but no minimum wage. Even the scope for market failure, therefore, the premise that the labour market might not actually clear in practice, had been rejected.80 Labour markets could operate competitively provided that the state did not “intervene”, and any attempt to do so by setting wage rate could not but lead to unemployment: “[payment of relief to] persons in full-time employment … is something which is bound profoundly to distort the wage system and to frustrate the ambition … that a man should receive as near as may be the full value of his work in cash”.81 The only exception to this was to be a limited system of wage-supplementation targeting low-wage-earners with dependent children.
By the mid-1990s, with no legally binding minimum wage and no comprehensive system of wage-supplementation to top-up wages to subsistence level, wage levels and benefit payments reached historically low levels.82 It was in this context that Labour launched its “New Welfare Contract”, a series of reforms intended to “make work pay” through a reformed tax-credit system, a central element of which was to be a statutory minimum wage. The minimum wage was not intended to be a living wage, however, as the Labour government made clear.83 The main purpose of the NMWA was to prevent employers responding to the tax-credit regime by decreasing wages, reducing incentives to work – anticipating, and correcting for, the distortive effects of a system of wage supplements. Making work pay was not to be confused, therefore, with “making employers pay”. This was not the function of the minimum wage84: “By setting a floor for wages, [the national minimum wage is] essential to ensure that low-income workers enjoy the full benefit of the tax credit.”85
The tax-credit system was expanded in 2002 with the introduction of the Working Tax Credit and the Child Tax Credit. In addition to providing a supplementary source of income to persons “engaged in qualifying remunerative work” this regime also made it possible for tax credits and other forms of income support (such as statutory sick pay) to continue during periods of illness, incapacity, maternity or adoption leave when the contract is continuing, but contractual pay (wages or sick pay) is no longer being paid.86 In this way, Labour's new tax-credit regime helped employers benefit from treating the employment contract as continuing when no work was being performed, without having to pay for it, relying on the State to support the worker until his return to work. By confining employers to paying the private costs of a commodity, it established an uncomfortable distinction between wages, and the costs of social reproduction.
This exclusion of the costs of reproduction from the concept of the wage is implicit in the structure of the National Minimum Wage Act and Regulations (NMWR) (explored in detail below).87 Like the Trade Board Acts, the focus of the legislation is on what the employer pays for labour in the market, rather than what the employee receives by way of remuneration for employment. Its primary purpose is to guarantee fair competition, shifting direct labour costs onto employers to prevent them from using low wages as an artificial subsidy to under-cut their competitors. The Low Pay Commission's remit does not extend, therefore, to considerations about the cost of living, confining them with considering the impact of any proposed rate on businesses and on the employment rate; on how much fairness the market could bear.88 To the extent that fairness could be considered, therefore, it was to be achieved within the constraints set by the market.
C. The Statutory National Living Wage
In 2012, the Coalition Government replaced the Working Tax Credit, and the five other existing means-tested benefits, with Universal Credit, to be phased in gradually.89 The tag-line for this policy was, once again, “to make work pay”, but the emphasis has shifted away from seeing tax credits as a mechanism for bridging the gap between wages and workers’ subsistence needs, towards a much tighter dove-tailing between tax credits and minimum wages to reduce dependency on all forms of state assistance. In this way, it intended to “remove barriers to employment, smooth transitions back to work and remove distortions in the current system that prevent people from doing extra hours when available”.90 It does this, however, by opening-up low-paid, insecure and part-time workers to a series of behavioural conditions that had previously been applicable only to the unemployed. To be entitled to state support, those in work as well as those out of work must meet a minimum (and progressive) earnings threshold, and must secure themselves additional hours and/or higher paid work/second jobs as a condition for claiming the credit.91 The focus of the “claimant's commitment”, the “contract” that claimants must sign before being given access to the scheme, is thus on requiring claimants to meet certain work-search, work-preparation, work-training and work-preparation requirements, to make it more likely that they will “obtain paid work (or more paid work or better paid work)” so as to avoid having their income reduced.
Intrinsic to this policy is the introduction, from 2016, of the National Living Wage (NLW).92 On its face, this would seem to be aimed at providing to all workers (over 25) a statutorily guaranteed “social wage” capable of meeting at least the basic costs of living.93 In practice, however, it falls short of doing so in many respects. Designed to work alongside Universal Credit's conditionality requirements, its principal purpose is to reduce benefit dependency by requiring workers to take on more hours and second jobs at the “NLW” level – without an express concern to tie that level to the costs of living. Instead, the wage is set to rise to a minimum rate of 60% of hourly median wages by 2020 – making no direct link between the wage and workers’ needs.94 The NLW is not conceptualised as a “living wage” in the sense implied by the Living Wage Foundation, therefore, an organisation that encourages employers to pay a voluntary “living wage” that is linked with the public's perception of what amounts to an adequate minimum standard of living.95 Whereas the “real” living wage attempts to make work pay by making employers pay a sum that reflects the social costs of labour, the NLW simply alters the distribution of those costs as between employers, and the state.
The National Minimum Wage Act 199896 already contained a power to pass regulations to provide for different rates of pay for those under twenty-six years of age. In fact, the 2015 regulations set out four minimum wage rates, corresponding with different age bands.97 Thus, when it came to implementing the proposals for the “national living wage” all that was required was a new “tier” to the age bands already provided in the 2015 regulations, renamed, however, as the “national living wage”. This is exactly what was achieved by the National Minimum Wage (Amendment) Regulations 2016, regulation 3 of which simply added regulation 4A to the 2015 regulations with a view to providing a wage supplement to workers aged twenty five and over.98 In practice, therefore, the introduction of the national living wage constituted a statement to the effect that workers below twenty five years of age were not entitled to be paid a “living wage”, while those above that age were entitled to be paid something that was, in practice, a market wage plus a wage-premium that would, in many cases, still fell far below the level that would be required to cover the costs of living.99
All the NLW does, therefore, is change the arithmetical relationship between the minimum wage and the tax-credit system – without touching the former's conceptual foundations. As will be shown below, no accompanying change was made to the structure of the NMWA and NMWR, nor the link between work and payment which they reflect. The policy is much more about encouraging workers to work as hard, and as much, as would be required at prevailing wage rates, than it is about seeing the satisfaction of the subsistence needs of the working classes as a conditioning factor on the rate of wages, and thus on the rate of profit.100
IV. The Conceptual Structure of the National Minimum Wage Act and Regulations
Most of the discussion in relation to the NMWA and associated regulations (NMWR) has concerned the meaning of the statutory concept of “work” as it is to this that the right to be paid attaches. However, this is itself of immediate relevance to the concept of the wage that the Act embodies. Tying the right to be paid to work demonstrates that the right to be paid a minimum wage is, today, a right to be paid a minimum market wage for labour.
The NMWA and NMWR draw a distinction between employment, referring to the existence of a contract (of employment or to personally provide work)101 that is a necessary condition for the application of the NMWA, and work, as that to which the right to be paid the minimum rate attaches.102 The scope of the Act is tied to employment, therefore, but the right to be paid the minimum wage is confined to time during that employment when work is being provided. It is not, therefore, a right to be remunerated for time employed.103 The scope of an employer's minimum wage obligations thus turns on how the concept of “work” is conceived and, because this concept is left un-defined, this will depend very much on the attitude of the courts.
The NMWR does not prescribe a particular mode of payment. It thus applies different principles to: time work, broadly corresponding with wage-work payable for hours worked; salaried work, applicable to those paid for a set basic number of hours a year; output work, such as commission and piece-rates; and “unmeasured work”, a residual category designed to catch all those whose working arrangements fall outside the other three categories. In each case, the obligation to pay is attached to “work”, subject to certain deeming provisions in Regulations 27 and 32, by virtue of which, salaried-hours workers and time workers also have a right to be paid the minimum wage for periods of “down-time” when the individual is “at or near” the place of work for the purposes of working.104 These deeming provisions do not provide a right to be paid for time spent waiting for work at home (neither “at or near” the workplace), therefore, time when the worker is “not awake for the purposes of working”, periods of leave that are expressly unpaid, and periods of industrial action.105
The notion of the “salary” to which the Act alludes is quite different from the historic idea of a salary as a fixed and unconditional payment for service over time.106 It applies to those whose contracts lay down a set number of basic hours for which they have, in theory, a right to be paid “regardless of the number of hours actually worked in a particular week or month”.107 It is not confined, therefore, to those who are employed in a particular capacity to provide their professional services. Moreover, because the courts defer to the individual contract to determine which of the worker's contractual obligations count as work, beyond the right to sickness and holiday pay, even those engaged in statutory salaried-hours work will only be paid for time when “work” is, or is deemed to have been, provided.108 For those whose contracts do in fact envisage ongoing service, therefore, the employer's minimum wage obligation will only extend to a fraction of the time that their contracts require them to devote to their employer's business (see below).
The category of “unmeasured work” reflects an alternative to salaried employment for employers wanting to benefit from having a worker on-hand to work and to do as much work, as and when required, without having to pay for the benefit of doing so. This category applies to workers whose hours of work are not specified, and who are required to work when needed or when work is available. Rather than being paid a fixed salary for time in employment, however, with the employer committing in advance to providing a minimum income or minimum hours of work, these workers are paid to complete a particular job or task, regardless of how long it takes them to do so and regardless of how (ir)regularly they are called upon to work. For minimum wage purposes, they must be paid for “the total of the number of hours spent … carrying out the contractual duties required”.109
Regulation 44 on unmeasured work operates in conjunction with regulation 49 which makes provision for the parties to agree a “daily average agreement” to determine “the daily average number of hours the worker is likely to spend in carrying out the duties required”.110 The problem with this provision is that if the worker agreed that she would normally be engaged in performing her duties for 12 hours a week, but the tasks prove so demanding (for reasons outside the worker's control) that she in fact performs closer to 20, she has effectively waived her entitlement to claim the minimum rate for the additional eight hours.111 The courts have said this provision is necessary, however, to avoid the “probable ending of [socially useful] employment … [by] pricing it out of the market”.112 Such an attitude has proven particularly problematic in the context of care-work, as the case of Walton illustrates well.
Walton v Independent Living Organisation (2003) involved a home-care worker providing a continuous service in the home of a severe epileptic. Walton claimed that she was paid by the day (“time work”) and engaged in time work throughout a daily period of 14 hours when she was awake and providing care and supervision. The Court of Appeal rejected the argument that all aspects of her activities during the day could be classed as work, classifying her work as unmeasured work. While the daily average agreement stated that Walton's caring tasks only consumed 6 hours and 50 minutes in any 24-hour period, she was required to be on the premises at all times in case the patient required assistance.113 Nonetheless, the court treated this as discretionary labour which did not attract the minimum wage, falling outside the tasks specified as work in the contract and extending beyond the hours agreed in the daily average agreement.
Walton involved employees engaged in care-work for what the courts classed as unmeasured work. Similar problems arise, however, for those engaged in time or salaried work whose contracts require that they remain responsive to the needs of clients/employers on an ongoing basis. In this context, the courts have tended to draw a “clear dichotomy” between those cases where a worker is working merely by being present at the employer's premises, whether or not they have been provided with sleeping accommodation, and those where the worker is provided with sleeping accommodation and is simply on-call (or available).114 In the former case, the worker will be treated as working throughout the full duration of the shift. In the latter, however, the deeming provisions in regulations 27 (salaried-hours work) and 32 (time work) will apply. This means that only the time when the worker is available to work “at or near” the workplace, and is “awake for the purposes of working” will be time to which the right to the minimum wage will attach.
If the contract clearly defines the work for which the employer is willing to pay, and requires in addition to that work that the worker perform a number of duties during “non-work” hours, the worker will have to rely upon the deeming provisions in order to claim a right to the minimum wage in respect of those additional hours. By careful drafting, therefore, the employer can narrowly circumscribe how much of the time from which he in fact benefits is time to which the obligation to pay the minimum rate will apply.115
This approach, excluding “non-core” work, as defined by the contract, from the “work” for which minimum wages can be claimed was doubted by Langstaff J. in Whittlestone (2014), and the EAT in Focus Care Agency (above).116 In the latter case, the EAT argued that the courts should apply a “multifactorial approach” to determining whether the worker is to be treated as working merely by being “available”.117 This requires a close examination of the parties’ contract, with particular emphasis on: (1) the employer's particular purpose in engaging the worker; (2) the extent to which the worker's activities are restricted by the requirement to be present and at the disposal of the employer; (3) the degree of responsibility undertaken by the worker; and (4) the immediacy of the requirement to provide services if something untoward occurs or an emergency arises.
This more “functional” approach is to be welcomed, but is limited because these factors seem only to be relevant where the contract does not itself clearly define what counts as work for the purposes of the right to be paid. For this reason, the “Focus Agency” test has been relied on to support the view that the courts must “start with the contract”.118 Rather than deriving any general principles, the EAT was thus content to leave each case to turn on the terms of the individual contract, including the way in which the contract provides for pay to be calculated. If the worker is in fact paid per shift, therefore, there will be a presumption that she is to be paid the minimum wage for the entire duration of each shift. If she is in fact paid by reference to tasks performed or time spent awake at the workplace, this would not be the case.119 The courts will not, in other words, impose an obligation to pay if it is not clear from the “broader factual matrix” that the time in question was something for which the employer was willing to pay.120 The EAT in Focus Agency endorsed the decision in Walton, therefore, suggesting that different principles are applicable to unmeasured work. There would seem to be little preventing an employer, particularly in the care-sector where on-call night shifts are common, from taking advantage of these provisions to further limit the scope of their minimum wage obligations.121
When Focus Care reached the Court of Appeal, the court overruled the EAT to the effect that sleep-in shifts will not be classed as “work” for the purposes of the regulations.122 Rather, Underhill J. argued that the deeming provisions were “self-evidently” intended to apply to sleep-in shifts, such that care workers engaged on such shifts were “available for work” and not “actually working”. Thus, the only time for which carers would be entitled to be paid were those hours when they were actually awake for the purposes of “working”:
the self-evident intention of the relevant provisions is to deal comprehensively with the position of sleep-in workers. The fact that their case is dealt with as part of the availability provisions necessarily means that the draftsman regarded them as being available for work rather than actually working. That is hardly surprising: it would not be a natural use of language, in a context which distinguishes between (actually) working and being available for work, to describe someone as “working” when they are positively expected to be asleep throughout all or most of the relevant period.123
Before the decision was released, Martin Green, the Chief Executive of Care England, stated that “if the existing decision of the Employment Appeal Tribunal is upheld it would be a watershed moment for the sector, with profound effects for the viability of residential domiciliary and supported care”.124 How far such concerns (and the scope for up to six years’ worth of back pay) influenced the Court of Appeal is an open question. It is clear from the judgment, however, that the EAT's multi-factorial approach was seen to go too far, to be an “unnecessarily elaborate approach” which is not necessary given the clear intentions behind the deeming provisions contained in the Regulations.125
This article explored the nature and function of minimum wage legislation by analysing how economic theories of the wage are reconstructed in juridical terms in the context of policy and legislation. By comparing two approaches to economic analysis, neoclassical economics, and classical political economy, it showed that how we understand the labour market, as a spontaneous order, on the one hand versus an instituted process, on the other, influences our conception of minimum wage and of the function of labour law in a market economy.
This approach made it possible to identify two competing conceptions of the wage in economic theory: the market wage, the price of a commodity, coordinating decision-making in the market; and the social wage, embodying fairness norms and customary living costs, channelling distributional conflicts, and underpinning the process of social reproduction. By drawing on the insights of the classical political economists, the article showed that there is a tension between these concepts, that a “social cost” problem is endemic to labour markets and can only be adequately addressed by institutional mechanisms, such as a minimum “living” wage.
The article went on to explore the conceptual structure of minimum wage legislation, tracing changes in the juridical conception of the wage over time. It showed that the economic categories of the “market” and “social” wage find their expression in the legal concepts of the “wage”, a payment for work that is linked to time spent working at the workplace, and “remuneration”, a contractual payment for service over time. Of course, the concepts used by legislators and the courts are not always consistent; but the analysis revealed that the concepts of the “wage” and “remuneration” have historically performed quite different functions in the context of the legal regulation of pay. For example, minimum “market” wages, as found in the Trade Board Act and NMWA, perform a narrow market correction function, while minimum “remuneration”, as found in the Wages Councils Acts, performs what has been referred to in this article as the social reproduction function, reflected in the notion of the “social wage”.
Since the 1980s, the policy underpinning minimum wage regulation in English law has been played out through the changing relationship between statutory minimum wages, and tax credits. This reflects a shift (back) towards a conception of the labour market as a spontaneous order and labour legislation something imposed from outside that affects its operation. In this framework, rather than providing for workers subsistence needs, the function of minimum wage legislation is simply to correct for specific market failure.
Between 1993 and 1998, following the abolition of the Wages Councils and in the absence of a statutory minimum wage, subsistence costs were to be borne not by employers, but by the state through a system of wage supplements and tax credits. The introduction of the statutory minimum in 1998 did little to change this: the minimum wage was aimed more at correcting market distortions than it was at shifting social costs onto employers: the tax-credit system continued to bear most of the weight when it came to topping up wages to subsistence level.
This article has shown how these ideas are reflected in the NMWA's conceptual structure, in particular in the distinction it draws between “employment”, as a condition for claiming the minimum rate, and “work” as that for which workers are entitled to be paid, and in the deference it affords to employers when it comes to their choice of payment system. Little attempt has been made to prevent employers from using the contract and its definition of “work” to shift economic risk onto workers.
The shift in policy from 1998 to today should not be under-estimated; while during both periods, the conception of the wage at the heart of the NMWA was a market wage that corresponded with the price of a commodity, prior to the more recent reforms, there was at least a tax-credit system in place that was expressly designed to top-up that wage to subsistence level. True, as the article has shown, this is itself not conducive to long-term stability and efficiency. But it does at least take seriously the need to provide for the subsistence needs of working households.
Rather than acting as a bridge between wages and workers’ subsistence needs, Universal Credit is more explicitly tailored to providing incentives to work – however irregular, short-term, or precarious that work may be. While the NLW does suggest that ideas about the social wage are resurfacing, the NLW as drafted is unlikely to be a true social wage, at least not for the majority of workers to which it applies. This is due to the wider changes being made to Universal Credit, further casualising work, the absence of any weekly or monthly minima, and the retention of the conceptual structure of the NMW within the shell of the new NLW. If the broader institutional environment remains hostile to stable employment, promoting casualisation, increases in the minimum rate alone will likely do little to help many of the lowest paid.
This article has revealed, however, that there exist a number of possible models for minimum “wage” legislation, and that there are more options available to us than simply “tinkering” with the “rate” at which the wage is paid. The analysis of the Wages Councils Act and the National Minimum Wage Act in particular indicates that the conceptual structure of a piece of labour legislation can play a role in what a given body of legislation can achieve. The premise that the wage should be paid for “work”, and that this expresses the limits of the worker's rights to be paid, is not, this article has shown, as self-evident as is often supposed. Rather, there exists a practical alternative to today's framework, based as it is on a link between work and wages, that might be better capable of providing protection against low and precarious pay; a model based on the principle of a minimum remuneration for employment.