首页 | 农村土地改革 | 法经济学茶座 | 法经济学论坛
     网站首页≯ 经济法≯
最低工资立法

MINIMUM WAGE LEGISLATION

Simon Deakin

Reader in Economic Law and Assistant Director, ESRC Centre for Business

Research, University of Cambridge

Frank Wilkinson

Senior Research Officer, Department of Applied Economics and ESRC

Centre for Business Research, University of Cambridge

© Copyright 1999 Simon Deakin and Frank Wilkinson

 

Abstract

 

Economists have traditionally been hostile to minimum wage legislation,

seeing it as an unwarranted interference with the operation of the market

and a cause of unemployment among the less skilled. Recent evidence from

the USA and the UK claims to show, however, that employment rises when

minimum wage laws are introduced and falls when such laws are repealed.

A number of explanations have been offered for these findings, ranging from

monopsony effects to labour market segmentation. There are trade-offs

between short-run adjustment costs and long-run improvements to

productivity and performance which are, however, difficult to assess. From

the point of view of the economics of law, the minimum wage studies show

how important empirical work is in testing and reshaping economic theory.

JEL classification: K31, J38

Keywords: Unemployment, Household Poverty

 

1. Introduction

 

Most developed economies (and many developing ones) set a legal minimum

level to wages, either through statute or by giving legal force to the terms of

collective agreements negotiated between employers and trade unions.

Economists, however, have traditionally been hostile to minimum wage

legislation and to labour standards more generally, seeing them as an

unwarranted interference with the operation of the market and a cause of

unemployment among the less skilled. Recent evidence from the USA and

the UK claims to show, however, that employment rises when minimum

wage laws are introduced (Card and Krueger, 1995) and falls when such

laws are repealed (Dickens et al., 1993). The story of how these findings

have been evaluated by the economics profession is of considerable interest

in itself for what it tells us about the present state of economic theory and the

way in which it intersects with the law.

 

2. The Traditional Neoclassical Account

 

In the traditional neoclassical account of the labour market, competition

between firms for labour, and between workers for jobs, ensures that wage

rates for labour of comparable productivity are more or less equal throughout

the market and beyond the power of any individual economic actor to affect.

The movement of the market towards equilibrium acts as an implicit

regulator of individual decisions on whether to trade and at what price.

Firms which attempt to pay below the market rate risk losing their workers

to competitors, in the same way that workers who attempt to force up wages

above the competitive level risk losing their jobs as firms at the margin

substitute labour for capital or cease to trade.

 

Repeated empirical studies - surveys and case-studies of firms dating

back to the first large-scale studies of low pay in Britain and the USA - have

shown that, in reality, labour markets do not display these characteristics

(see Nolan, 1983). Firms do not automatically adjust wages to changes in

demand for labour, and there is considerable divergence in the pay and

conditions offered by different employers to workers doing similar jobs.

Neoclassical theory tries to explain away these findings by the argument that

freely competitive markets tend towards equilibrium. Limitations upon the

flow of information and upon the mobility and substitutability of factors of

production are overcome in time by the unravelling of competitive market

forces. The market itself is seen as a powerful force for equality: under

conditions of perfect competition, equal pay for work of equal value would

follow from the operation of supply and demand. Where apparent

inequalities in pay persist, they are viewed as the result of pre-market

factors: differences between individuals in terms of endowment and ability,

differences in individuals’ ‘tastes’ for work, training or leisure, or

differences in employers’ ‘tastes’ for discrimination (Becker, 1957).

 

Regulation is also seen as an exogenous cause of inequality, preventing

market clearing. Public choice theory regards labour legislation as the

outcome of organised pressure-group activity (Heldman, Bennett and

Johnson, 1981; Rowley, 1985). Trade unions, according to this point of

view, are labour monopolists seeking to cartelise the labour market. By

driving wages above the market or equilibrium rate, they depress demand for

employment and divert resources into wasteful rent-seeking. The costs of

trade union activity are borne by consumers, in the form of artificially high

prices, and the unemployed, who are ‘priced out’ of work. Trade union

legislation, by supporting collective bargaining and the right to strike,

ensures that these inefficient monopoly practices are protected against the

market forces which would otherwise drive them out (Posner, 1984). On

similar grounds, it is predicted that minimum wage laws would have a

disproportionately adverse impact on the young, those without formal

training or qualifications, and those re-entering the labour market after a

long absence (such as the long-term unemployed) (Minford, 1985).

 

3. Empirical Evidence

 

The minimum wage has given rise to a vast empirical literature. Until

recently, the consensus was that the introduction of a minimum wage and

subsequent rises in its level would both almost certainly lead to increases in

youth unemployment, and sometimes to increases in adult unemployment

(Brown, Gilroy and Kohen, 1982). However, this conclusion rested mainly

upon time-series studies using long-term aggregate data of teenage

unemployment derived from a single source, the US Current Population

Survey. Critics suggested that these studies could not be regarded as

definitive since the estimated employment effects are small and also highly

sensitive to the choice of sample period (Card, 1991).

 

Case studies by Card, Katz and Krueger examining the implementation

of minimum wage reforms in various US states in the late 1980s and early

1990s presented a different picture (see Card, 1991; Card, Katz and

Krueger, 1993; Card and Krueger, 1995). The research took advantage of

the opportunity for comparative study which arose from the variations in

rates of increase between states, and by the decision of the US Congress in

the late 1980s to raise the federal minimum wage after a period of several

years when its nominal value remained constant and its real value declined.

One study examined the effects of raising the minimum wage in California

in 1987, comparing teenage employment rates with those in states which did

not increase their minima at the same time. It was found that both the

earnings and the employment of teenagers in California increased after the

minimum wage was raised, despite over half the teenage employees in the

state being affected, a very high compliance rate, and few exemptions being

allowed in the legislation. Similarly, a study comparing New Jersey, which

increased its minimum wage, with Pennsylvania, which did not, found

evidence of increasing employment in the former state. An analysis of the

implementation of changes in the federal law in the fast food industry in

Texas, a sector employing mostly part-time workers and with labour

turnover rates in excess of 300 percent per annum, found that most

employers observed the new adult rates for the minimum wage but that few

took advantage of the possibility of paying a teenage sub-minimum. Over 70

percent of the firms interviewed did not report responding to the new rates

by either dismissing workers or cutting fringe benefits.

 

In Britain during the same period, minimum wage regulation was being

weakened as a consequence of legislation which cut back the powers of the

Wages Councils (wage-setting bodies operating at industry level in certain

sectors). After 1986 they were prevented from setting rates for younger

workers and in 1993 their powers to set wages and conditions were

completely removed. Prior to the 1986 Act it was ‘confidently postulated’

that the abolition of the Wages Councils would ‘serve to expand employment

[and] offer competitive wages for the socially disadvantaged’ (Minford,

1985, p. 122). However, econometric studies found that employment in

low-paying service sectors declined as a result of the decreasing effectiveness

of the Wages Councils (Machin and Manning, 1994; Dickens et al., 1993).

In a similar vein, studies of the UK Equal Pay Act 1970 found that while the

legislation was responsible for a considerable narrowing of the male-female

gap, this was accompanied without a fall in female employment levels

(Zabalza and Tzannatos, 1985).

 

4. Alternative Theoretical Perspectives

 

The US findings on the positive impact upon employment of minimum wage

increases produced a stormy debate. Some critics argued against the use of a

comparative case study methodology, while others sought to provide

different explanations for the observed employment effects (see Keenan,

1995; Neumark and Wascher, 1995). However, when the findings are taken

together with the British research which demonstrates the converse effect -

the negative consequences of weakening minimum wage enforcement - they

can be seen to have far-reaching implications for the conventional

understanding of how labour market regulation works. Under these

circumstances, it is not surprising that attention has focused on alternative

theoretical perspectives.

 

One possibility which is compatible with orthodox neoclassical theory is

that employers in low-wage sectors are monopsonists holding a degree of

market power over their employees. This enables them to keep their wages

below the equilibrium or external market rate. It would be in the interests of

such employers to avoid taking on new employees at the market rate if they

then felt obliged, on the grounds of equity, to raise the wages of their

existing employees. Under such conditions, a limited rise in the minimum

wage would necessarily lead to a rise in employment, as the higher wages

attracted new recruits into the industry concerned. Too large a rise, on the

other hand, would lead to unemployment, particularly where there were

wide variations in the degree of monopsony power which individual firms

possessed (Stigler, 1946).

 

A more radical departure from the orthodox account is provided by

theories of labour market segmentation. These focus on the link between low

pay and industrial structure, and take a historical and institutional

perspective which is to some degree compatible with theories of path

dependence or institutional lock-in. It is argued that since low pay is

concentrated in particular occupations and industries, pay inequality cannot

be wholly explained by differences in the skill and quality of individual

workers (Craig et al., 1982). Rather, low pay comes to be associated with

jobs which are socially undervalued or which are performed by workers who

are accorded a low labour market status. ‘Undervaluation’ is a consequence

of a number of factors including employer strategies, differences in the

effectiveness of worker organisation, and the division of labour within the

household. For example, one effect of the sexual division of labour is that

skills traditionally associated with women’s non-waged labour in the home,

such as caring and cleaning, command lower pay in the labour market than

other comparable ‘male’ skills. Linking women’s pay to skill levels is, in

practice, highly problematic. Studies comparing male and female earnings

across a range of occupations have suggested that as much as three quarters

of the wage gap between men and women in comparable jobs could not be

put down to skill-related factors (Horrell, Burchell and Rubery, 1989).

 

A labour market segmentation perspective leads to a very different view

of minimum wage legislation (Deakin and Wilkinson, 1992). Rather than it

being an ‘artificial’ interference in the free market, it becomes just one form

of regulation which, together with other conventions, norms and customary

practices, governs the way in which labour is contracted. The case for

legislation is that as a consequence of segmentation, certain groups in the

labour market will not have access to voluntary means of labour

organisation, such as collective bargaining or the protection of professional

rules governing entry and access to jobs. Hence the industries in which low

pay is endemic are those in which there are structural factors, such as ease of

entry by both firms (the result of low capital requirements) and workers (the

result of the failure to develop and enforce formal skills and qualification

requirements), which impede the effective organisation of labour and hence

the application of common terms and conditions. Although these sectors

appear more than any other to resemble the neoclassical ‘norm’ of free

competition, in fact they are the exception. The persistence of these

conditions creates the basis for a fundamental disequilibrium: low pay is in

effect a subsidy, enabling otherwise uncompetitive firms and industries to

survive. Minimum wage regulation is therefore necessary in order to help

create an environment in which firms compete not on the basis of low pay

but instead through high labour quality and product and process innovation

(Deakin and Wilkinson, 1999). Equally, placing a floor under wages can

augment the purchasing power of workers and so underpin effective demand

(Michie, 1987; Prasch, 1996). These macroeconomic effects have indeed

been prominent among the justifications offered for the introduction of

minimum wage regulation in most of the systems adopting it.

 

5. Dynamic Effects of Introducing and Raising the Minimum Wage

 

The general case for retaining a minimum wage, if one is in place, is distinct

from more specific arguments for and against introducing or reintroducing

minimum wage regulation. A fundamental change of this kind in the

regulatory framework may well have far-reaching effects which cannot be

precisely predicted. Moreover, it is unlikely that the effects will all be in one

direction. Although an argument can be made, as we have just seen, for

long-term efficiencies arising from the presence of a statutory wage floor,

these may have to be traded off against short-run adjustment costs as some

enterprises go out of businesses and workers retrain. Other matters to be

decided include the level at which the minimum wage is set, and the

mechanism through which it is subsequently varied.

 

Many of these issues have been the focus of intense debate in Britain

since the election of a Labour government in 1997 which was committed to

reintroducing minimum wage regulation after an interval of four years

during which no minimum wages of any kind were in operation. The

government-appointed Low Pay Commission (LPC), which reported in June

1998, recommended the introduction of a national minimum wage

(previously minimum rates only had the force of law in certain industries)

(LPC, 1998). The proposed rate was towards the lower end of the range put

forward by advocates of the minimum wage, namely £3.60 per hour from

April 1999, rising to £3.70 in June 2000. The £3.60 figure represented about

45 percent of median earnings and was expected to affect about 11 percent

of the working population over the age of 20. Sixteen and 17 year olds were

to be exempt and 18-20 year olds were to receive a lower rate of £3.20 per

hour from April 1999, rising to £3.30 in June 2000. The Commission

estimated that the effect of introducing these reforms would be to add 0.6

percent to the total national wage bill; the sectors most heavily affected

would be cleaning, catering and security, while the groups to benefit most

included women workers and homeworkers. The government accepted the

Commission’s recommendations for adult workers but set the starting rate

for 18-20 year olds at £3.00 per hour, rising to £3.20 by June 2000.

 

The work of the UK Low Pay Commission represents an interesting

example of the use of economic evidence to aid public policy. The

Commission was sympathetic to the monopsony model, noting its

implication that ‘moderate minima can be introduced without destroying

jobs’ (LPC, 1998, p. 114), while at the same time noting the broader

‘undervaluation’ argument in particular as it applied to female labour (LPC,

1998, p. 115). It also considered that the introduction of a minimum wage

could have an impact on industrial structure:

 

Whatever the nature of the labour market, it is likely that a National Minimum

Wage will have a greater effect on the structure of employment than on its level.

Businesses which are inefficient or which produce low value-added goods may

 

need to reorganise working practices. If the National Minimum Wage is properly

enforced, business and employment are likely to transfer to more efficient firms

or to those offering higher value-added products and services … minimum wages

may cause a transfer of jobs between groups such as the substitution of more

skilled for less skilled workers … . (LPC, 1998, 115)

 

It is debatable whether the level set by the Low Pay Commission is

sufficiently high to realise these potentially beneficial effects of

reintroducing a minimum wage. A minimum wage set at about 45 percent of

median earnings is low by international standards (the minimum wage in

France, for example, is 57 percent of the median), although it is higher than

the US federal minimum. In setting the rate at a low level, the Commission

may have minimised the short-run adjustment costs of low-paying

employers, at the expense of forfeiting a longer run improvement in

productivity and economic performance from this source.

 

A significant feature of the new UK legislation is that no provision is

made for automatic increases of the minimum wage in future years. In the

USA, a similar lack of automatic uprating has led to stagnation and decline

in the level of the federal minimum, and has turned the level of the

minimum wage into a hotly debated political issue, so arguably increasing

the scope for wasteful pressure group activity (Sachdev and Wilkinson,

1999). In France, by contrast, legislation embodies a formula whereby the

‘minimum growth wage’, the salaire minimum interprofessionel de

croissance or SMIC, rises automatically with prices and with at least half

the increase in the purchasing power of the average wage. This has ensured

that, alone of minimum wage systems in western Europe, the SMIC has

retained its real value since the mid-1980s (OECD, 1998, chapter 2).

 

6. The Minimum Wage, Household Poverty and Work Incentives

 

The minimum wage, it is sometimes said, disproportionately benefits

households with more than one person in employment, some of whom are

so-called ‘secondary’ earners (typically wives and children of a male

‘breadwinner’). In practice, the employment of ‘secondary’ earners is often

essential to improving the living standards of multiple earner households. A

more important objection is that minimum wage appears at first sight to

offer no means of relieving poverty in households with no member in

employment. In practice, however, the minimum wage may assist the

unemployed by improving work incentives. This depends upon how the

minimum wage interacts with the benefits system. The contribution of the

minimum wage to alleviating poverty and inequality needs to be set in the

context of other, complementary measures, in particular those operating

through the social security system.

 

The institutional choice here is whether to have a system based on

wage-substitutes, such as an extended negative income tax, or whether to use

the minimum wage in conjunction with social security benefits to provide

incentives for the non-employed to re-enter the labour market at a wage level

which makes it worth their while to work. In Britain, the former system was

in operation between 1986 and 1997. Social security benefits were targeted

on low-paid earners with families at the same time as the minimum wage

was abolished. Because these benefits were (unavoidably) means tested, they

were progressively withdrawn as earnings from wages increased, so

extending the ‘poverty trap’ which reduces the incentive for workers to take

better paid jobs or jobs involving longer hours (Deakin and Wilkinson,

1991).

 

Moreover, since employers could reduce wage levels or at least avoid

increasing them in the knowledge that the difference would be met by social

security, the state was in a position of subsidising low-paying firms which

were often among the least efficient in terms of productivity. This

reinvention of the Speenhamland system (see Polanyi, 1944) soon led to

huge increases in state expenditure on benefits for the low paid.

 

After 1997, the election of a new government led to a change in

emphasis. Benefits to the low paid were retained but restructured with the

intention of building on the level set by the minimum wage rather than

directly substituting for it. As part of these reforms, taxes and social security

contributions for the lower paid were substantially reduced. The aim of the

reforms was to ‘demonstrate the rewards of work over welfare’ (Treasury,

1998).

 

7. Conclusion

 

The recent debate over the minimum wage demonstrates the remarkably

tenacious hold which orthodox neoclassical theory has over the economics

profession, even when confronted with evidence which, to say the least, puts

in doubt the predictive capacity of that theory. From the point of view of the

economics of law, the minimum wage studies show how important empirical

work is in testing and reshaping economic theory. The hostility of

mainstream law and economics towards labour regulation, as exemplified by

the symposium on labour market published in the University of Chicago

Law Review in 1984, is just one of many instances in which apparently

clear-cut normative conclusions were drawn from models which had only a

weak link to real-world conditions. From an institutional perspective such as

that which can be drawn from the theory of labour market segmentation, the

effects of the minimum wage - in particular, the consequences of introducing

or raising the minimum - can be seen to be highly complex. There are

trade-offs between short-run adjustment costs and long-run improvements to

productivity and performance which are, however, difficult to assess. This

may lead us to conclude that while the economic analysis of labour standards

may help improve our understanding of how such laws operate, in and of

itself it does not provide clear normative guidance to policy makers. At the

end of the day, the case for social policy interventions will continue to be

made on broader, ethical grounds.

 


中国法律经济学网登载此文出于学术研究之目的,绝不意味着中国法律经济学网赞同其观点或证实其描述。以上内容仅供研究者学习与交流,无意侵犯版权。如有侵犯您的利益,请告知。我们将尽快删除。

加入日期:2007/1/10 20:45:41浏览次数:2573
发表评论
名号:
内容:
验证: 3900
法律经济学网
联系站长: 柯华庆 lawgame@263.net 京ICP备09028584号
北京市昌平区中国政法大学法学院(102249)
本网站由卡卡鱼网提供技术支持 网站总访问量:1996356