., y s 0 c I E .......................................................... FABIAN PAMPHLET 545 Making a Minimum Wage Work • by Fred Bayliss .......................................................... FABIAN PAMPHLET 545 Making aminimum wage wor1< Introduction 1 1. Origins of the proposal 2 2. Purpose and scope 5 3. The machinery 9 4. Economic effects 14 5. Practical steps to a NLMW 18 \ Fred Bayliss is chairman of the Campaign for Work. He is the author ofBritish Wages Councils. Now retired from the Department of Employment, he was formerly at the NBPI, the CIR, the Pay Board and the Royal Commission on the Distribution oflncome and Wealth This pamphlet, like all publications of the Fabian Society, represents not the collective views of the Society but only the views of the authors. The responsibility of the Society is limited to approving its publications as worthy of consideration within the Labour movement. Design: Tony Garrett May 1991 (ISBN 0 TI63 0545 3 ISSN 0307 7523 Printed by The CollegeHill Press Limited (TU), London and Worthing Published by the Fabian Society, 11 Dartmouth St, London SW1H 9BN 21 0037738 4 TELEPEN Introduction I BLPES I The Labour Party and the TUC have been committed to a national legal minimum wage (NLMW) since 1986. The commitment was in the Party's manifesto for the last general election, and will be in the manifesto for the next one. The recent policy document Opportunity Britain promises that 'side by side with our National Economic Assessment, we will introduce a national legal minimum wage, starting at a level of 50 per cent of median men's earnings. Over time, we will increase the minimum wage as a proportion ofearnings to a point where no-one is paid less than two thirds of the median male hourly rate'. This pamphlet is about the implementation of a NLMW. Turning the proposal into a practical reality will not be easy. Its simplicity is one of its chief attractions -an hourly rate of pay below which no one can be legally employed. But the significance of a NLMW for pay determination generally and its impact on the conduct of employers and trade unions are complicated. Unless they are fully taken into account a NLMW could be a failure. 1 1Origins of the proposal There has been legal intervention in wages for many years. Wages Councils go back to 1909 and the extension of them to catering, retailing and other trades after 1943 was one of Ernest Bevin's most enduring achievements as Minister of Labour. H H owever, the Councils have been in decline since the 1960s both in the numbers they cover and in the significance of their Orders. Moreover, after 1979 the Conservative government put the abolition of Wages Councils on its agenda. Sir Geoffrey Howe asserted in his 1981 Budget speech that 'the machinery ofWages Councils has operated to price people out ofjobs'. By 1985 the Government had decided not to abolish but to emasculate them. In the 1986 Wages Act employees under 21, some of whom were among those most in need of protection, were excluded from the coverage of the Councils, which were also confined to fixing a single rate of pay with an overtime premium for those aged 21 and over. Had the government followed the logic ofits conviction that legal minimum wages always reduce employment significantly, it would have abolished them. But ministers were deterred from doing so by the opposition of employers' bodies who feared the unfair competition ofunorganised employers and by the argument that abolition would strengthen the case for a NLMW. In 1990 they again backed off abolishing the Councils. But the possibility of abolition and the actual reduction of the Councils' powers pulled the rug from under what support for Wages Councils remained among the unions. The alternative-a NI:MW-had to be actively pursued. The question of a NLMW was not new to the TUC. It had been considered and rejected in 1970. During the 1970s the TUC had a policy on low pay of wage targets for negotiators which was particularly associated with Jack Jones, General Secretary of the TGWU. The debate for and against a NLMW was first thoroughly aired at the 1974 Congress on a NUPE motion in favour ofit.The motion was defeated. The General Council opposed it on the grounds that most unions preferred low pay to be dealt with through collective bargaining supported by a target wage, regularly updated. But as Alan Fisher, General Secretary of NUPE, said in the debate, some workers covered by 2 collective agreements were among the low paid, as well as many who were not. That was to be the central issue of the debate for the next twelve years-was collective bargaining capable ofimproving the position ofthe low paid or had there to be resort to the law? The main reason why the TUC took so long to come down in favour of a NLMW was the opposition of the TGWU. Given its size and influence in the TUC, it was unusual for a major policy to be adopted with it in the minority, and similarly in the Labour Party. It leaders consistently argued that 'the way forward for the low paid workers is to join trade unions' and 'the first priority must be to organise the unorganised'. A NLMW would hold back the pay of others by having 'a drag anchor effect'. Employers would use the NLMW as a 'lever to depress wage levels' and 'it could even be the slippery slope to a statutory incomes policy'. However, NUPE skilfully gathered support and this powerful traditional argument was eventually rejected. In 1983 ASTMS, a union with little direct interest in a legal minimum wage, announced its conversion. In the following year the National Union ofTailors and Garment Workers, almost all ofwhose members were covered by Wages Councils, switched sides saying that a NLMW might help its members to achieve a wage two-thirds of national average earnings, which negotiations never would. The Low Pay Unit pursued a persuasive campaign in the unions and the Labour Party. The majority votes of the TUC and the Labour Party conferences in 1986 signalled the victory of those who wanted a NLMW. In July 1987, after the General Election, the TGWU biennial delegate conference came down in favour ofa NLMw which 'provided a safety net for the low paid and permitted free collective bargaining' but which was not part of'an all-embracing statutory pay policy' The decision ofthe TUC in favour ofa NLMW was part ofits general move towards greater reliance on the law which was a feature of the mid-and late 1980s. Statutory provision for the recognition of unions by employers, also adopted by the Labour Party, was the other main example. Nevertheless, the implementation ofa NLMW must not ignore either the potential which union organisation and collective bargaining still have for raising the earnings ofthe low paid or the hesitation of some unions to having a minimum wage set by law. The inadequacies ofWages Councils Wages Councils have been the established method for dealing with low pay for many years. Their central feature has always been that they set statutory minimum remuneration by industry. This was regarded as desirable because it replicates collective bargaining. The 'side' members of Councils are mostly representatives of employers' organisations and unions in the trades even if 3 they have only a few members. The independent members see their function as the narrowing of the gap between the two sides when a pay claim is submitted and ifthat is not possible they vote with one side or the other. Until 1986 the Councils fixed different legal rates for different grades ofworkers as well as hours, overtime rates, shift premia and holidays. Their Orders looked like collective agreements although everything in them had statutory force. The Councils were a system for providing substitute collective bargaining in a collection of trades where pay was low. They used to be described as 'infant' collective bargaining to bring out the purpose of their growing up into 'adult', independent, negotiating bodies. But employers' organisations and trade unions often came to regard them as virtually permanent. However, there were a few cases where the Councils grew up and were abolished on the joint application ofemployers and trade unions. In the early 1970s the Commission on Industrial Relations recommended the abolition ofa number on the grounds that very few employees had earnings ofless than the statutory minima plus 10 per cent. In all, 21 Councils were abolished in the 1960s and 1970s. So by 1988 the Council system had been reduced to 26 Councils covering 2 million employees. At their peak the Councils had covered 3 million employees. The Councils are not capable of producing a legal minimum wage with comprehensive coverage. They are not even comprehensive in the coverage of their own trades. The two retail Councils, for example, do not cover pharmacists, florists, butchers, dry cleaners or betting and photographic shops. The catering Councils do not cover guest and boarding houses. There are no Councils for trades like contract cleaning, which have large groups oflow paid workers. And the small numbers of low paid workers found in every trade, including those covered by collective agreements, who in total number several hundred thousand, many of them part-time workers, are also outside the Councils. The Councils are not directed to improve the relative pay ofthe low paid nor are their procedures suited to such an objective. The Councils' procedures steer them towards following other settlements.Itis not surprising that the historical record shows that the Councils have not improved the relative position of the low paid. The system had been in decline for twenty years when the government emasculated the Councils in 1986. They can no longer be justified as 'infant' collective bargaining since they are now confined to setting only one rate of pay.The low paid in other trades have no protection. The Councils are neither one thing nor the other. They are no longer supposed to lead to independent negotiations nor are they a viable method ofsetting a general legal minimum wage. 4 Purpose and scope 2 The Labour Party and the TUC have turned to a NLMW because collective bargaining supplemented by wages councils has failed to improve the relative position of the low paid. T T he expansion of service industries and the spread of part-time employment, meanwhile, have increased the numbers vulnerab.le to low pay. The purpose of a NLMW is to make a significant and permanent ili!pact-~Il. the r.oblem of ~-y-j;-;;;;-be gi;enamore, or a less, ambitious form. The more ambitious purpose is to raise the earning ofthe low pa1d relative to those of others and so make the distribution;£pay_ less unequal. The less ambitious purpose is to use it as a safety net through ~one falls so that the relative position of the low paid does not get worse. The more ambitious purpose fits in with a wider political programme designed to reduce differences in income from work and social security benefits, after tax.The less ambitious purpose represents continuity with attempts to ensure that the low paid keep step with rates in collective agreements and acts as a safety net. Although this pamphlet takes the more ambitious line one of its main themes is the difficulty of achieving an improvement in the relative position of the low paid. In practice the safety net may be all that a NLMW would be allowed to achieve. Within that broad, more ambitious, purpose there are two major and two minor objectives. Most of those who stand to benefit from a NLMW are part-time women working in service trades where there is no representation by trade unions. They are the lowest paid because their position in the labour market makes it easy for employers to take advantage of them. They often have family responsibilities which limit the hours they can work, they cannot travel far, they have few skills or cannot get jobs which use the skills they have. They have to take what jobs they can get. Employers seldom take them into jobs requiring skills so they have little training and few prospects of moving into better jobs or of being promoted. They are treated as marginal to the main labour market. Union membership is not on the cards. These are workers who are without protection. 5 Some examples bring it home. Women going, say, from Brixton to the City at 6am or 6pm to do three hours or so office cleaning five days a week are 'invisible' in more ways than one. They work when others do not, they probably only know their supervisor who may have several buildings to look after, they travel and work when their familes are at home, they often do this work because they cannot get anything else, and ifthey did not take the pay offered they would probably have no job. Other examples can be found among staff working on cleaning and kitchen duties in hospitals, or bar staffworking split shifts at week-ends, or shop assistants doing a few hours in the morning for sole traders. I I The main objective of a NLMW is to provide protection for those whose ' market power is least and who are at the mercy of employers who take advantage oftheir weak situation. A NLMW should be seen as an instrument of public policy designed to compensate, in part, for their exposed position in the labour market. In some ways it is analagous to the legislation on health and safety. The consumer can offer no protection. The employers feel that competition hardens their treatment of the low paid, and unions are many miles away. Only the power of the state through the law and its enforcement agencies can give them protection. The second major objective is to make that protection effective. The NLMW should make the low paid better paid relative to others by ensuring that others do not take action to restore differentials and generally build on top ofa NLMW so that its effect is eroded. Also, its purchasing power should not be eaten away by inflation. Such objectives are difficult to sustain absolutely and 'as far as possible' has reasonably to be added. There are two other objectives which flow from these. A NLMW should contribute to reductions in poverty. Low pay is not coterminous with poverty. Many people in poverty do not work. Many low paid people are not in family poverty. Poverty in families headed by a wage earner depends on the number ofdependants as well as the level of payjl'he wage set by a NLMW would be hourly and income from employment is governed by the number of hours worked as well the hourly rate of pay. But there is an overlap of poverty with low pay, particularly for families. Taking low pay as 50 per cent of median male earnings and the poverty line as the long-term rate ofSupplementary Benefit, the 1983 Family Expenditure Survey data showed that nearly 10 per cent offamilies in poverty were headed by a low paid wage ear~er (l)..tkaising the earnings of the low paid would reduce the number offamilies m poverty. But there are many in poverty who would not be helped by a NLMW, just as there are many who would benefit who are not in poverty. That is why a NLMW cannot have the reduction of poverty as a main objective. It can make some difference but its contribution should not be exaggerated and it should not divert attention from the other and more important weapons available for use against poverty. 6 There is a connection between low pay and social security benefits which is a fourth objective. Broadly, unemployment benefit is less than wages otherwise it would usually make sense to take the benefit and not work. So benefit is, in a way, kept down by what the unemployed would eam in jobs. The interaction of benefit and wages partly controls what can be done about poverty.1;Raising low levels of pay should make it possible to raise benefit. Indeed, 'a NLMW, by setting an across-the-board minimum level of pay, simplifies the lowest wage to be compared with benefit. The development of the connection between the minimum wage and unemployment benefit in order to raise benefit as a proportion of average pay should be deliberately fostered. Potential coverage The best indicator of what constitutes low pay is the level of earnings of all full-time employees below which 10 per cent fall (the lowest decile). It cannot be a target figure but it is a pointer to the eamings below which recipients are low paid. That figure in April 1990 in the Department of Employment New Earnings Survey (NES) was 327p (60 per cent of the median earnings of all full-time employees) and it indicates that about 1/4 million full-time and 2 million part-time employees, 4 million in total (about 20 per cent of all employees), were low paid. That is the size ofthe problem. Part-time workers are about 60 per cent of the low paid. Over 80 per cent of the low paid are women and nearly 60 per cent of the full-time and 80 per cent of the part-time low paid are concentrated in distribution, hotels and catering, and other services like hairdressing and hospitals. The low paid are mainly women working part-time in service trades. The Labour Party's proposal sets two targets for the level ofa NLMW-half median male hourly earnings at the outset and eventually two thirds of the median male hourly rate. In April 1990 median male hourly earnings were 596p, so half was 298p. About 7 per cent of all full-time workers and 34 per cent of part-time workers eamed less than 298p. There is no distribution of hourly rates of pay so it is not possible to know how many employees have earnings ofless than two thirds ofthe median male hourly rate. In its place, two thirds of median male hourly earnings are used in this pamphlet (although that figure, 399p, is bound to be somewhat larger than two thirds ofthe median hourly rate). 23 per cent offull-time and 69 per cent ofpart-time employees eamed less than399p in April1990.Turning those proportions into numbers of employees gives the following: in 1990 3 million workers eamed an hourly rate of less than half the median (of which 1.2 million were full-time and 1.9 million part-time). 7.3 million workers eamed an hourly rate of less than two thirds the median (3.4 million full time, 3.8 million part time. -f 7 These are large numbers and indicate that about.,l in 7 o all employees would be affected by a NLMW pitched at earnings ofhalfthe median for men and abo~in...3 empl(}yees-aHwo thirds. The proportions affected would be greater in regions where earnings generally are low. For example, at halfthe median in the Northern and East Midlands Regions and inWales one full-time woman employee in six and more than two part-time women employees in five would qualify whereas in the South East those proportions would be only one in about twenty and one in I ~0 , five. All Wages Council hourly rates current in April1990 were well under half median male earnings of 298p. The retailing Councils were on 269p, the catering Councils between 233 and 250p. Outside London, the bottom two grades oflocal authority manual workers (284 and 296p) and the bottom four scales ofthe Health Service ancillary staff(262 to 297p) fell below halfmedian male earnings, as did the lowest rates in some private sector agreements, such as textiles. So the Labour Party lower target would overtake not only all Wages Council hourly rates but also the main rates in local authorities and the NHs·. The Labour Party proposal uses halfthe median offull-time men's earni s as a target. But a NLMW would apply to all employees irrespective of sex indicating that half the median earnings of all full-time workers is more directly relevant. In April1990 that figure was 275 . ince part-timers were incompletely covered by the NES their earnings are excluded from the calculation ofthe summary figures and it is not possible to know the median hourly earnings of all employees, full-time and part-time together. But there is a strong case for using 275p as the target, ifthere is to be a target, which would produce 1.8m qualifying compared with over 3 million at 298p in the table. 275p in April 1990 was higher than all Wages Council rates, the two lowest rates in the NHS and a number ofthe lowest rates in textiles but in all cases, of course, by a smaller margin than 298p. Such wide variations in estimates of the numbers qualifying and in the trades affected using different versions ofthe same definition indicate the deceptiveness ofsuch an apparently simple target. Notes. 1. Bazen S: Minimum Wage Legislation, the likely impact on earnings in the UK LSE 1988. 8 The machinery 3 The objectives set out in the previous chapter suggest certain features of the machinery to give effect to a NLMW. I I f a NLMW is to be the protective arm of the state for the low paid the machinery must be placed firmly in the government's domain. The level ofthe NLMW has to be seen to flow from the state and that means from the law and the government ofthe day. The main purposes ofthe NLMW must therefore be set down in legislation with ministers being made responsible for arrangements to carry out the law's provisions, including the fixing of the level of the NLMW in statutory orders. The process for setting and re~sing the NLMW must_take account of all the factors which affect whether the minimum wage sticks. That involves making informed assessments of how employers and trade unions are likely to react to it in the pay levels they negotiate. Changes in the structure of 1) employment as they affect the numbers of the low paid have to be looked t. D Eshmates have to be made of the likely 'employment and price effect at different levels of the NLMW. The ·n:ilwith-liiieiDPloyment be efit_has to be kept in mind. Looking at all these factors, a balance has to be struck between pushing the legal minimum wage up as fast as Qossible while at the same time not provoking such reactions as to endanger the real advance made by the low paid. This all suggests the need for a Minimum Wage Commission which could take all these matters into consideration and propose a level ofthe NLMW to Ministers. The Commission's terms of reference should be spelt out in a Schedule to the Act so that it would not be a creature of any government. Its composition should be determined by the NLMW's main objectives. It should hav~ members whose knowledge and experience equip them to design the state's protection for the low paid. Some should be knowledgeable about the circumstances ofthe low paid at work and in society. Others should be expert in the working of the labour market and, in particular, of systems of pay determination. There should also be places for representatives of employers' bodies and trade unions, who can contribute on what can be expected from their constituents by way ofreaction to various levels of a NLMW. The Commission would need to make itself well informed about the extent oflow pay and the effects ofthe NLMW. It is important that it should do that 9 in ways which show its commitment to the interests of the low paid. That probably means that asking organisations for their views is less important than direct inquiries and surveys and it may be that the Office of Manpower Economics could provide these Eervices to the Commission. Ministers and the Commission have to knit their functions closely together. Ideally, the Commission's recommendations on the level of the NLMW would be adopted by ministers and implemented. But the inherent tendency would probably be for the Commission to wantto move faster than ministers. Itmight tend to err on the side ofbelieving that employers and unions would not make agreements which prevented the low paid catching up. Ministers would tend to be more hesitant on that score, unless there were a framework for coordinating pay movements, including the NLMW. A deliberate concentration of the annual pay round into the period between a National Economic Assessment in the autumn and the Budget in the spring would improve the prospects ofsuccess for the NLMW especially ifthe NLMW were decided after the main pay negotiations had been concluded, as has been suggested by John Edmonds and Alan Tuffin (1). Setting the NLMW at the same time as the National Economic Assessment and before the main agreements are negotiated, as the Labour Party currently proposes runs the risk ofnegotiators using the NLMW as a springboard. The initial level ofthe NLMW Setting the NLMW for the first time is a different exercise from revising the level subsequently. Special arrangements would need-to be made for the initial process, especially when it would involve the appointment of members of the Commission and the first interpretation of its terms of reference. The Commission should carry out a thorough review and report on all the relevant factors so that there could be a full debate before it took the step of recommending the initial level of the NLMW. Ifa bad mistake is made first time round it will be difficult to put right later. Setting the initial level is the only occasion when the Commission will not be preoccupied with the size of the increase in the NLMW. The response of employers and unions to the initallevel is likely to be at its most unrestrained. They will compare the initial level with the lowest hourly rates ofpay in their agreements and focus on the the narrowing ofdifferentials. They will be aware that their reactions will set a limit to the responses they might make to later revisions. The joint statement of the Labour Party and the TUC in 1986 contained wise advice on this issue: 'The important point is not the initial figure; it is to get the legislation on the statute book and progressively up-date it'. It is the level achieved in the long-term, not the short-term, which matters. Nor should it be regarded as immutable that the first level has to be half median male 10 hourly earnings or any other target; the initial level should be left to the Commission to recommend and the government to decide. An annual cycle should be adopted for the NLMW. The Commission should make an annual report setting out in detail its assessment with a recommendation, preferably atthe end ofthe main annual pay round. Butifits protection of the low paid is to mean anything, there should be something close to a guarantee ofthe NLMW's real value. No special protection is required. By at least keeping pace with median earnings the NLMW would rise fast enough since most increases in pay which feed the increase in medi~arrungs meet- inflation. Price indexation could not, in any case, be confined to the NLMW. Even so the Commission's terms of reference should make it clear that there is a presumption in favour of protecting the real value of the NLMW. Ifthe relative position of the low paid is to be improved, the NLMW must move up by bigger increases than pay generally. That is the central issue for the Commission because over a period of years it needs to be ahead of other pay movements and there will, inevitably, be a tendency for others always to want at least as big an increase as the low paid. The French system (SMIC) In Europe the French system is the most relevant to what British practice might become (2). In Germany, Denmark and Italy wage rates set in collective agreements are extended by law to low paid workers and others whose employers are not directly covered by them. The other countries in the EC (apart from. Ireland whj.ch has the equivalent of Wages Councils) have a NLMW. In France there is an annual review when the government receives the opinion ofa joint employer and employee advisory committee. Before 1970 account was not taken ofthe general rise ofearnings but since then the review must ensure that 'in no case may the annual increase in the purchasing power ofthe SMIC be less than halfthe growth in the purchasing power ofthe mean hourly earnings'.(1970 Statute) So halfthe growth rate ofreal hourly earnings generally is guaranteed. Ifbetween reviews prices rise by more than 2 per cent an automatic trigger brings the SMIC fully into line with prices. Moreover, the government can make special additions to the SMIC at its discretion as it did in 1981 and again in 1989 and 1990. In these ways the system ensures that the real value ofthe SMIC never falls by more than 2 per cent, that a real increase is guaranteed if average real wages are rising, and that the government can push the SMIC up even faster. France is the only EC country where price indexation is automatic. When the SMIC's predecessor was introduced in 1950 there were no guarantees on prices and real earnings; they were added to an established system in 1970 and at the same time regional minima were abolished. A further development in 1988 was the creation oflinks between the SMIC and social security benefits. As a result of developments in the legal minimum 11 wage over forty years the SMIC is now fully integrated into the working ofthe French labour market and poverty programmes. Taking the 1980s as a whole the SMIC has been rising as a proportion of average earnings. It shows that the more ambitious purpose can be achieved. Enforcement There is no point in having a NLMW which employers can avoid and on which employees cannot rely. It is a commonplace of enforcement that self-policing is the best. There are some aspects ofa NLMW which work in that direction. It should be comprehensive, covering every employee aged 18 and overinevery place of employment, in public as well as private establishments, and with virtually no exceptions. Its comprehensiveness would be enhanced if 16 and 17 year olds were entitled to percentages of it. Every form of employment should be covered with no exceptions for, say, casual workers or personal servants like nannies and houseworkers. Only the severely disabled, covered by registration, should be exempt. This simplicity would have great advantages for enforcement. A comprehensive NLMW with one figure could be given widespread and easily understood publicity. Notices should be displayed in all public places and the minimum hourly rate should be extensively advertised so that it would soon be know to everyone. It should be presented as every citizen's right. The word 'minimum' should be emphasised so that it is clear that no one is precluded from offering or from requesting more. Inspection would be easy because it would only be necessary to check the rate of pay. Since no employer could plead ignorance of what should be paid the threat of prosecution and stiff fines for failure to pay could become a powerful deterrent. But there would need to be extra effort at the outset in order to ensure that the initial level was universally applied.The presentWages Inspectorate,even ifconsiderably strengthened, would be like a drop in a bucket. What is needed is a short cut to quick detection of failure to pay in the first few months. The VAT inspectors regularly visit establishments in the trades in the private sector where most low paid people who stand to benefit from a NLMW are employed. They could be required to do a quick check in every establishment they visit and to inform the Wages Inspectorate of any prima facie infringe ments. In the long run the best way to enforcement is people's knowled~e of their right to the NLMW and a speedy response to any complaint. But it would need to be backed up in a number of ways. Complaints should be facilitated by telephone help lines and complainants should be protected against being sacked by the extension of the unfair dismissal procedure. Unions and indi viduals should be able to take speedy action in the courts. The penalties 12 should include compensation for the people underpaid on top of the actual arrears. There is also the possibility of evasion. The more effective a NLMW the greaterthe temptation for some employers to pay less. Two main escape routes would have to be policed. Casual work for unrecorded cash in hand would be illegal if it were below the legal minimum hourly rate. Self employment on terms worse than the NLMW would be illegal ifitwere disguised employment. Ifit were legitimate self-employment the income could legally be less than the NLMW. In France there is virtually complete observance ofthe SMIC. It is simple and universal and everyone is used to it. For example, there is no hesitation about offering the SMIC in advertisements and about one fifth of employees in small firms are on it. Simplicity and universality are possible in Britain from the start but the absence of experience and understanding mean that a special effort on enforcement must be made when a NLMW is introduced. Notes 1. John Edmonds and Alan Tuffin: A New Agenda· Bargaining for Prosperity in the 1990s. GMB!UCW 1990. 2. See Mark Minford: Minimum Wages in Europe. Low Pay Review, Summer 1989. 13 4 Economic effects The NLMW is not a magic wand and cannot be considered in isolation. I I t will raise the rate ofpay which employers mustoffer to those previously paid less. It will affect em lo ers' costs rices and rofits and cause them to look at the n hers they employ and the ways in which their ~ductivity can be increased. There Wl be consequences for the settlement ofthe pay ofworkers other than those whose pay has been increased by the NLMW. Within firms differentials will be reduced at the bottom of the wage structure as the lowest rates are overtaken by the NLMW and some restoration is likely. Wage claims put in by unions, some in industries with few low paid workers, may seek to use the NLMW as a starting point for the increases they wan hese economic effects are part and parcel of having a NLMW. The main qtiestion is whether they can be controlled and prevented from reducing the benefit of a NLMW to the low paid. A NLMW would have a direct impact on labou osts. This would vary greatly between industries according to the numbers who had to be brought up to its level. Using the yardstick of a NLMW fixed at about half average earnings the direct effect on the labour costs of all industries and services taken together would be less than 1 per cent. But in those trades where the low paid are concentrated the increase in labour costs would be much greater -retailing 3 per cent, clothing, hotels and catering 7 per cent.ft\lhere low pay is particularly extensive the increases in labour costs would be much greater; in hairdressing, for example, it may be as much as 20 per cent. At two-thirds ofaverage earnings the impact ofa NLMW on overall average labour costs would·be over 3 per cent and in those particular industries and services mentioned in the previous paragraph it would be two or three times the effect at half average earnings (1). Some other estimates using different sources yield higher figures but the high impact in service trades and the low overall effect is not in doubt. These direct cost effects would be repeated to some degree each time the NLMW was revised and increased by more than the average of other pay movements. Of course labour costs are not the only costs, or even the most important costs, in many industries, so the effect on total costs would be much less. But in addition the knock-on effects dealt with below would cause an indirect increase in costs. The impact of these labour cost increases on prices would be small on average but in some service trades it would be considerable. What would shop 14 keepers, hoteliers, publicans, restaurateurs and hairdressers do ifthey had to pay a NLMW which increased labour costs significantly? They could do four things, or a combination ofthem.They could put up prices to recoup the higher wages (which is a solution not always to be objected to since customers ought to be willing to pay prices which cover a legal minimum wage). But having put up prices they may well lose custom and have to reduce their operations and dismiss staff. They may, however, respond by putting in more capital equipment or improving management so that the productivity of their employees rises and unit costs and prices do not increase. Finally, they may take a cut in profits. The increase in labour costs in the public sector becomes an increase in government expenditure if no steps are taken to absorb it through greater efficiency. However, few of the low paid in this sector earn significantly less than half average hourly earnings. A NLMW at that level would add slightly over 1 per cent to the pay bill in education and the NHS (estimated to be £392 million in 1986) although it would rise to 4 or 5 per cent at two-thirds average earnings. Knock-on effects A NLMW could have consequential effects on payinthree ways. As has already been noted, it would have a small effect on prices since direct labour costs overall would rise by less than 1 per cent. Another source would be the restoration ofthe differentials reduced by the NLMW, in whole or part. In the trades most affected, this knock-on effect would depend on the characteristics ofthe pay structure and who determines it. In a service trade where the structure is only sketchy and there is a large gap between the rates at the bottom overtaken by a NLMW and those higher up, and where there is no collective bargaining, there would probably be few consequential pay increases. On the other hand, in public sector areas such as hospitals and local government, where a NLMW would overtake the lowest rates of pay in a closely packed wage structure controlled by collective bargaining, there would be strong pressure to restore at least part ofthe narrowed differentials. The more the restoration was by the use of tapering increases the smaller the knock-on effect. Going beyond the trades most affected, it may be that in others, although few people would get increases out of a NLMW, the lowest rates in the wage structure would be overtaken. Despite its small significance in particular places of work, unions may nevertheless seek restoration of the differentials all the way up the structure and that would result in serious knock-on effects. In the TUC debate in 1986 about a NLMW the EETPU put down a hostile motion and Eric Hammond, its General Secretary, said that a statutory minimum wage 'which seeks to erode differentials is neither suitable nor 15 practicable'. In the same debate, Bill Jordan, President of the AEU, while supporting a NLMW warned that 'there would be a relentless battle to restore differentials at a stroke (which) are an accepted recognition ofskills acquired and we will defend that principle'. Very few members ofeither union work in establishments where the NLMW would affect more than a handful of employees so the restoration of differentials would depend not on direct experience ofthe low paid catching up but on attitudes to low paid workers in distant trades improving their relative pay. A general restoration of differentials would virtually put a ban on the low paid improving their relative position. The third way is the possibility that the increase in the NLMW may be regarded by unions and employers as indicating a new 'going' rates. Such a read-across from the NLMW would be very damaging. It would maximise the inflationary effects of the NLMW through pay negotiations across the economy. In France they have the opposite problem. The SMIC has overtaken low rates of pay without the structures of pay rates being jacked up as a result, although there is some evidence that it has fed across into earnings. It is now a matter ofconcern to the French government that the bottom rungs ofmany wage structures have disappeared as the SMIC has advanced and it is disinclined to use its discretion again to improve the SMIC until these wage structures have been reformed. The success of the SMIC at improving the relative position of the low paid owes a great deal to the absence of knock-on effects. Two conclusions can be drawn. There are both specific and and general effects of a NLMW capable of setting in train reactions which would cut into the benefit which the low paid are intended to derive from a NLMW and also into the prospects ofnon-inflationary pay settlements. But these undesireable consequences are not remorseless and inevitable. They depend on how employers and unions decide to react. If a NLMW were a fixed proportion of median earnings, like half, the restoration of differentials would have to be prevented, otherwise the proportion could not be held. If differentials were restored to any degree, median earnings would rise and that would necessitate a compensating increase in the NLMW, setting off more restorations of differentials. An automatic upward chase would have been set in motion with inflationary consequences. Since there will in practice be some restoration of 11 differentials, the NLMW should not be a fixed proportion ofmedian earnings. The second conclusion is that as far as possible a NLMW should be set and increased in ways which provoke the minimum ofoff-setting responses. That means not setting it too high and not moving it up too fast. What constitutes 'too high' and 'too fast' would have to be explored by the Commission with representatives of employers and unions. IdentifYing the level of the NLMW which would set off the minimum of consequential changes and secures the maximum gain for the low paid is the key. 16 Employment effects All these effects can be summarised in terms of the reduction in employment which could be expected when the NLMW is introduced. After taking into account various levels ofthe likely cost and knock-on effects, the employment effect has been estimated using different models of the economy. Taking account of a number of such estimates an informed judgement is that 'no concrete answer can be given' to the question of what the employment effect would be: they suggest 'upper bounds of about 250,000 and 880,000 for the employment effect ofintroducing a NLMW set at half the male median wage and two thirds of the median respectively; how much less depends on the full economy-wide effects of the NLMW. It is quite probable that the effect is less than halfthese 'upper limits' (2). It is clear that the reduction in employment consequential on a NLMW cannot be brushed aside. But it is also clear that the employment effect of a NLMW at any level could be reduced considerably ifthe reactions ofemployers and trade unions were on the following lines. In establishments where a NLMW has a direct impact on the pay of a significant number of employees steps should be taken to improve productivity in order to facilitate the absorption of the higher labour costs and the minimum of restoration of differentials should occur. In those industries and establishments where few or none of the employees got increases from a NLMW and there was almost no direct impact, there should be no reaction at all because no differentials in places ofwork had been upset. At around halfmedian male earnings the economic effects ofa NLMW are containable ifthey are minimised by taking appropriate action. On then other hand, if the consequences are simply left to work themselves out they could be destructive both ofthe gains intended for the low paid and for employment. The cost and employment effects at two thirds median earnings (and presumably at two thirds the median male hourly rate) are significant. Even if that level is only approached over a period of years, minimising the knock-on effects may require a more concerted policy by employers and unions over several pay rounds than they are capable ofsustaining. Notes 1. See Brosnan P and Wilkinson F: A National Statutory Minimum Wage and Economic Efficiency. Contributions to Political Economy 1988. 2. Bazen S: On the Employment Effects of Introducing a National Minimum Wage in the UK British Journal of Industrial Relations. 1990. 17 5 Practical steps to a m1n1mum wage The implementation of the NLMW should be dominated by the objective ofbenefitting the low paid. Plainly it would not be to their benefit if the NLMW were set at a level which caused other groups to claim more as a result, or which led to GOnsiderably more unemployment among the low paid themselves .. T T he significance of a NLMW should not be exaggerated and its prospects should not be worsened by over-ambitious targets. Low pay is part of all pay. In Britain decentralised collective bargaining is the main method of settling pay and so there are many points at which pay settlements could be made which would erode the intended benefit of a NLMW to the low paid. It will not be easy to strike the right balance between preferential treatment for the low paid and the negotiating objectives of those with bargaining power. But unless it is recognised that the main danger to the effectiveness ofthe NLMW is posed by those who are better paid, the low paid will be in for disappointment. The first job is to get a NLMW introduced. That means legislation and the initial report by the Minimum Wage Commission. Getting a NLMW in place should be seen as more important than any particular starting level. A modest level with the minimum ofreactions should be followed by increases as big as the Commission judge can be absorbed without provoking action which stops the low paid improving their relative position permanently. Keeping those objectives in mind the main practical steps which can be recommended are the following: • The NLMW must be comprehensive. Total coverage of employees with the minimum of exceptions would play a major part in getting the NLMW accepted when it is first introduced. Any loop holes would be enlarged as time went by. Its credibility requires that there be 18 no room for argument. Every person over 18 should be legally entitled to at least the NLMW hourly rate ofpay. Even an exception for the disabled should be tightly drawn and regulated. 16 and 17 year olds should have percentages ofthe minimum. Deductions are a potential source ofweakness and should be strictly confined, if possible, to the provision of accomodation at stipulated rates. • A Minimum Wage Commission should be set up. Its members should include people knowledgeable about low pay and pay determination generally, and representatives ofemployers' bodies and trade unions. The legislation should be drafted so that it gives the Commission considerable independence as well as defining the powers of Ministers. It should be the Commission's business to investigate and report on all aspects of the operation of the NLMW and to take all relevant factors into account when advising the government on the level to be set and on revisions ofit. The Commission would require work to be carried out for it and it may be that it could be done as part of the duties ofthe Office of Manpower Economics. • The government should lay down the NLMW. Ifthe Commission does its job properly Ministers will usually have no difficulty accepting its recommendations. Both the use of the law and the wider relevance of the NLMW mean that the government must hold the power of decision. In particular, Ministers should take steps to integrate the NLMW and the benefit system. • The NLMW should rise faster than average earnings. If the low paid are to catch up and improve their relative position a NLMW must increase faster than the average pay of others. That is ambitious and means that employers and trade unions must be willing to let the low paid advance. A less ambitious objective would be to use a NLMW as a safety net to prevent the position of the low paid getting worse. • Pay negotiators should not take action which reduces the benefit of the NLMW to the low paid. The Labour Party and TUC joint statement in July 1986 recognised the dangers. It said, 'unions will be expected to undertake not to quote in claims for higher paid workers that element of general percentage increases in earnings specifically related to the general move to obtain the national minimum wage' and 'it is impossible to eradicate low pay without destructive inflationary consequences unless .. higher earners receive smaller increases than the low paid'. Those are clear, unambiguous and forthright guidelines 19 and they would need to be observed by employers as well as unions ifthe low paid are to receive the intended benefit ofa NLMW and inflationary repercussions are to be avoided. • Specific target levels for the NLMW should be avoided. Ifthe Commission is to take account ofthe reactions ofemployers and unions to the level ofthe NLMW it must have scope for deciding at what level erosion would be at a minimum. The case for an increase in a NLMW is likely to be the result of a number of factors; mandatory targets would militate against that approach. There can be no guarantee that any particular level will be proofagainst being wom away. Indeed, an attempt to commit the Commission and the govemment to a specific level may well be the best way of making certain that reactions are unrestrained. That is not to say that there cannot be aspirations about what should be possible. Halfmedian hourly eamings of all workers is practicable. However, long-term aspirations should not be formulated until there has been some experience of the NLMW. The Labour Party's commitment-to halfmedian male eamings as the initial level and two thirds of median male hourly rate eventually -makes it less likely that the early years ofthe NLMW will see improvements in the relative position ofthe low paid. • Action should be taken to reduce the effects of higher costs. Even if action damaging to its beneficial effect is minimal, the NLMW would still cause increased wage costs where the low paid are concentrated with the possibility ofunemployment. Those effects can be tempered ifemployers take steps to increase productivity and so absorb the higher labour costs. They could be encouraged do so by a programme of government grants for advice from consultants to improve the utilisation ofmanpower, by analogy with the DTI's Enterprise Initiative. ACAS has a network of contacts with firms and their employees through its regionally based advisory services, and it could perhaps be the agency for administering the grants. In the guidance given by the Secretary ofState for Employment to TEC's for the content of their plans there could be a requirement that training in the trades where the low paid are concentrated should be given priority. • Enforcement must be taken seriously Ifthe NLMW is simple and comprehensive it stands a good chance of policing itself eventually. But in the early days it must be evident that the penalties on employers who fail to pay it are certain and severe. There would be no plausibility in any pleas of ignorance. The Wages Inspectorate's job would be made easier by there being only a single hourly rate to enforce but ifmore than 2 million workers are affected at the outset even a larger Inspectorate would still not convince employers that infringements are likely to be discovered and 20 punished. Some short cut is needed. VAT inspectors go regularly to most establishments with employees where the NLMW would be lifting wage rates. They could be required to do a check of wage payments as a matter of course and inform the Wages Inspectorate of any prima facie infringements. In that way confidence in enforcement which is essential to the effectiveness of the NLMW at the outset would be greatly improved. In the longer term complaints would need to be supported by protection against dismissal, by access by unions and individuals to the courts, and by compensation for individuals who had been underpaid in addition to the payment of arrears of pay. • Wages Councils should be abolished. The Labour Party proposals say nothing about Wages Councils, leaving open the possibility that they will continue to exist. Those who support the retention of the Councils want them to be able to build industry minimum rates on top ofthe general NLMW. Ifthat were so, the existing Councils with their present powers could only set another single legal minimum rate. Itwould be illogical to have two legal minimum rates in some industries but not others, the Council rate always presumably superseding the NLMW. It would be possible would be to run a NLMW side by side with Councils which possessed the powers they had prior to 1986 to set separate legal minimum rates for all occupations, together with overtime and shift premia and holidays. But that would raise new inconsistencies. Councils cover only about half of the low paid and it would not be defensible to have only some of the low paid covered by Councils as well as a NLMW. Consistency would require that all had both and that would mean a Council for virtually every industry where there was inadequate collective bargaining. That would seem to be defeating the purpose of introducing a comprehensive NLMW. Universality is worth having and a NLMW is the simplest way of getting it. The introduction of a NLMW means abolishing Wages Councils. 21 Making a minimum wage work ......................................................... The Labour Party is committed to introducing a national legal minimum wage. But will this help the low paid, and what will be the wider effects on employment, inflation and competitiveness? Fred Bayliss, chair of the Campaign for Work and former member of the NBPI and Pay Board, argues that the introduction of a minimum rate of half median male earnings would affect 3 million people, one in seven employees. A rate of two thirds median earnings would affect 7.3 million, about a third of the total workforce. He proposes a Minimum Wage Commission composed oflabour market specialists, economists and representatives of business and the trade unions to advise on the level, but argues that the final decision must be made by the Government. Enforcement would be facilitated by widespread publicity and a complaints procedure. VAT inspectors could be required to check wages and report any prima facie infringements ofthe minimum wage. He concludes that a minimum wage of half median earnings would be practicable without serious adverse effects, but that an attempta to move beyond could spark off a reaction, making it less likely that the relative position of the low paid will improve, and more likely that destructive economic consequences will ensue. - £3.50 The Fabian Society brings together those who wish to relate democratic socialism to practical plans for building a better society in a changing world. It is affiliated to the Labour Party, and anyone who is eligible for membership of the Labour Party can join; others may become associate members. For details ofFabian·membership, publications and activities, write to: Simon Crine, General Secretary, Fabian Society, 11 Dartmouth St, London SWlH 9BN.