2000 words economics essay on wages regulation.
johnathan1993
wage_floors-4apr15.docxForthcoming in Chris Coyne and Rachel Coyne – “Floors and Ceilings - the economics of price controls”, London: IEA Hobart Paperback.
The Simple Economics of Wage Floors
W S Siebert January 2015
Birmingham University
ABSTRACT
The paper surveys minimum wages setting and finds adverse employment effects broadly in line with conventional economic models. UK studies generally report small effects, but the UK is poorly suited for measurement given confounding changes in welfare benefits and no regional variation. Clarity is found in better empirical settings such as Canada, with good minimum wage variations and provincial differences, and in the OECD country panels. Case studies are also reviewed including South Africa (minimum wages as a white supremacy weapon), Portugal (extended collective agreements being analogous to minimum wages, and reducing employment), and Greece (the minimum encouraging temporary work). Alternative ways of helping the poor via encouragement of stable families, more competitive school systems, and subsidies for low paid work are advanced.
“My daughter’s ambition is to get a job in an office. She has Down’s syndrome. She thinks that, if she works hard, someone, somewhere will give her a job. At £6.50 per hour, it’s never going to happen. But at £2 per hour? Maybe.” (Letter to Daily Telegraph 17/1/0/14, from Candice Baxter)
“There is now no sizeable lobby in the UK campaigning for the abolition of the minimum wage….In a poll of experts by the Institute for Government the minimum wage was voted the most successful UK government policy of the past 30 years, ahead of the Northern Ireland peace process” (Manning 2013, 65)
Introduction
A wage floor such as the minimum wage makes payment of low wages illegal. Such a floor clearly tends to reduce unskilled job opportunities, yet it is only one example of floors under working conditions placed by regulation. Other floors on terms and conditions of employment relate to requirements for protection against unfair dismissal, or against discrimination, or for the provision of pensions via“autoenrolment”. Moreover, we must remember that high welfare benefits also place a type of floor under wages, since for many it is not worth working for a wage lower than the welfare payments they can receive. The adverse effects of these floors can compound each other, particularly in a high tax environment, as we will show. High floors can also be imposed by union power, especially via extended collective agreements as in France.
The minimum wage from the beginning has been justified by the Low Pay Commission (e.g. LPC 2000, para 3.18) as a means of achieving “equity in the workplace”. But, in most private sector businesses, equity is already achieved, in the sense that wages approximate the revenue product of the marginal worker. Private sector competition drives this result though the public sector of course does not fit this model so easily. If low wages are made illegal, then what happens is that the least productive workers cannot be employed. This result is demonstrated most clearly in the case of disabled workers, as shown in our opening quotation above. As Candice Baxter points out in her letter, her daughter could gain employment at £2 per hour, but certainly not at £6.50. One’s heart goes out to her. The celebrations of the politicians in the Institute for Government, shown in our second quotation from Professor Manning, are premature.
Wage floors and other regulations of working conditions grow together with centralised government, and are a part of the EU dirigiste tradition. The 1989 EU Charter on Fundamental Social Rights of Workers Rights marks a watershed (see Addison and Siebert, 1994), and has subsequently become the Social Chapter of the 1992 Treaty of Maastricht and 1997 Treaty of Amsterdam. The Social Chapter sets floors to most aspects of employment conditions, including “fair remuneration”, working hours, freedom of association/unionisation, training, equal treatment for men and women (and others), compulsory worker consultative councils, and health and safety. In EU terms (Commission 2006, p5), “the purpose of labour law is to offset the inherent economic and social inequality within the employment relationship”. In other words, decent wages and conditions are due to the efforts of politicians. Thus, the role of free markets and freedom of movement in defending the under-privileged is misunderstood.
Setting minimum wages is easy, but deals with the symptoms of low pay, not the causes. The political payoff from minimum wage laws is immediate. The dispersion of wages is reduced, and, since more women are low paid, so is the difference between male and female average pay (another misleading statistic). Yet nothing is done about the real problems in the labour market and the education system (see Kristian Niemietz 2012). The low level of skills acquired by children from our many single parent families is ignored, as is the worklessness among these families (see below)[footnoteRef:1]. As for really disadvantaged groups such as the disabled, the minimum wage may do much harm. Opportunities may also be reduced for students who are prevented from taking low paid internships (and may have to volunteer instead), and for those whose main work is in the home, but who would like to obtain some work to supplement household income or gain the benefits of socialising at work. The best that can be said of the minimum wage policy is that it is irrelevant to real problems of inequality and worklessness. More likely, it is part of a package combined with other floors on working conditions, strong unions and high taxes which make matters worse – as exemplified by Greece. [1: 26% of dependent children aged 0-18 live in single-parent families in the UK, which is almost twice as high as in France and Germany. The proportion of children being brought up in jobless families is consequently also high, around 20% (see OECD 2011a Tables 1.1 and LMF1.1A) which reduces these children’s education and employment prospects.]
Our plan is next to consider research into the UK’s national minimum wage, which is difficult given the lack of regional variation and the confounding effects of high levels of welfare payments. Then, we will take up results from Canada and the US where provincial variation in the minimum gives a more suitable design for minimum wage evaluation. We will also discuss the interesting case of South Africa where minimum wages were for a time used as a weapon in the struggle for white supremacy. Here we will also discuss results from long OECD country panels, which arguably give the best design for minimum wage evaluation. Finally we will extend the discussion to consider effects of minimum wages set by collective agreement. Such minima are more detailed and intrusive as shown by Martins’ (2014) study of the “30,000 minimum wages” set by collective agreements in Portugal.
UK Evidence on Employment Effects
The UK is the worst place conceivable to test for minimum wage employment effects. Changes in the minimum have been quite small, they are country-wide (so there is no clear counterfactual), and they are carefully tailored to the unemployment situation so as not to exacerabate unemployment unduly. (The economists on the Low Pay Commission are apolitical, and well aware of negative employment effects). Compounding the problem is the changing welfare system which also affects employment. We should remember that in 1999, at the same time as the minimum wage was implemented, the government introduced Working Family Tax Credits designed to encourage work which would obviously tend to counteract minimum wage effects in the opposite direction. However, the UK research does need to be considered, if only to show that we need to be careful before concluding (see for example Leonard et al 2014) from small measured UK minimum wage effects that conventional labour market models do not work.
The confounding effect of movements in welfare entitlements is shown in Figure 1. Here simple demand (D) and supply (S) curves are drawn for the unskilled labour market. If a minimum wage is imposed, equilibrium moves from point E to point A. However, if welfare benefits are brought in, or raised above the minimum, then employment falls further, to point B. If welfare benefits are pre-existing, then the impact of the minimum wage on employment will be muted. The diagram is a simplification, because apprentices and trainees will continue working for less than welfare benefits, as we discuss later. Also, welfare benefits – certainly in the UK – may vary with income from employment, so that some people receive some benefits even if they are earning a wage below the welfare floor. But the tendency remains – there is an interaction between the welfare system and the minimum wage.. Moreover, if welfare benefits are reduced, or reformed via tax credits conditional on work, employment will increase, confusing minimum wage effects.
Table 1 shows welfare trends, including the important housing assistance component. Unfortunately this series on net replacement rates is not available prior to 2001, however we can analyse most of the period since the introduction of the minimum wage. As can be seen, replacement rates have declined in France and Germany (the Hartz reforms), increased for single people in the US, and increased for families but not for single people in the UK. In fact, the UK’s family replacement rates are now amongst the most generous in the OECD (Niemietz 2012, p46). However, some of the welfare is contingent on working at least 16 hours, and so encourages some work. In fact, Gregg et al (2012, p22) estimate that employment of single women rose by 4% when tax credits were introduced: this worked against minimum wage effects oon employment.
Figure 2 demonstrates two important points about the progress of wages after the minimum was introduced. First, we see that a substantial number of people were freely working for low wages in 1998 prior to the implementation of the minimum. About 6% of the workforce is in that lower tail which grades slowly down to zero, reflecting the alternatives and productivity levels of the individuals concerned. Here Mrs Baxter’s daughter would have found her job.
Secondly, we see that even after the minimum was introduced, many wish to be paid below it. Low pay is a continuing phenomenon, with currently about 200,000 adults (21 and over, i.e. 1% of the adult workforce) paid below the minimum (LPC 2014, p133). In fact, the real total is probably double this figure (Kay 2007, p35) when we add the large number of under 21s. Much of this non-compliance occurs with apprentices, who are quite happy to receive low wages while training, as do students in general. In fact, about 70% of 18-20 year old hairdressing apprentices (LPC 2014, p134) currently refuse to accept the minimum wage, for which of course the Low Pay Commission and trade unions wrongly blame the employer. A further group are the interns and volunteers, who do not enter wage distributions like Figure 1 at all, but whom the Commission sees as unfairly avoiding its system. The important point is that many people see it worthwhile to work for low or no wages. This is particularly true for students and others training in order to raise their future wages.[footnoteRef:2] [2: See Gorry [2013, p72) for a good paper on how the minimum wage obstructs training: “Inexperienced workers are unable to pay for their training through reductions in their wages. To gain experience, they must maintain employment in a segment of the labor market characterized by high job separation rates” – their probability of unemployment consequently rises.]
A good study of adverse employment effects is provided by Dickens et al (2012), who focus on part-time women where coverage by the minimum is about 10%, twice as high as for full-timers (LPC 2014, Fig 2.1). They establish a counterfactual by comparing part-timers whose wages are raised by the minimum to those paid just above that level, i.e. 10% above. This comparison group will have similar skills and welfare benefit options. They find that the introduction of the minimum in 1999 caused the year-on-year probability of part-time women retaining a job to fall from around 0.70 to 0.65. Another way of putting this finding is that before the minimum wage, median job duration was about 1.9 years for part-time women earning around the minimum, falling to 1.7 years after the minimum, that is, a fall of around 10%. They do not find effects caused by the up-ratings as they have been too small to have much effect. For example, the recent up-rating from £6.31 to £6.50 gives a maximum uplift of 1.5% to about 5% of the workforce. As well as being small relative to welfare benefit changes, minimum wage increases have been lower in times of recession. Overall, this means that it is difficult to detect effects of changes in the minimum wage, but this result does not mean that higher increases in the future will be harmless.
While minimum wage increases so far have been small, a further piece of research by Rebecca Riley (2013) brings out the negative effects of the altogether larger change that the introduction of a minimum wage equal to the “living wage” would have (see also Siebert 2014). Currently the “living wage” is calculated to be £7.65 outside London, and £8.80 in London (Living Wage “Commission”, 2014). Such a change would mean wage increases for about 25% of the workforce rather than 5%, and raise labour costs of young unskilled (non-university) workers by as much as 14% in sectors such as hotels and catering, and retailing. Riley shows not only that the demand elasticity for labour is negative, but also that cross-price elasticities are generally positive, precisely in accordance with conventional economic theory. In other words, when the wage of the young inexperienced and unskilled group goes up, their employment falls, and the employment of substitutes such as educated workers and older workers increases. Thus, 300,000 young unskilled workers will lose their jobs, but some skilled and older workers will gain, so the overall loss of jobs is reduced to 160,000. Increasing the minimum wage to the living wage change would thus enable the older, better educated workers to gain at the expense of unskilled youth, as happens in France, but of course with serious long-term consequences for those trapped outside the labour market.
The current dark situation for Britain’s unskilled youth is shown in Tables 2 and 3, covering the period since the minimum wage began. Table 2 shows that labourforce participation has declined for all the disadvantaged groups except the disabled (whose participation remains low). Admittedly it appears that our youth participation rate is still better than that in France. But the UK lags on another measure which is shown in Table 3, the percent of 15-19s in the not in education, employment or training (NEET) group. Here we see that the UK’s percentage has been growing and at 9.5 % is the worst of the five major economies shown, and worse than the OECD average. In summary, Tables 2 and 3 show that the inequality of life chances has been growing, despite the Low Pay Commission’s mission to reduce “inequity in the workplace”.
People who reject the orthodox explanation for the small UK minimum wage effects need an alternative, They bring forward ideas of single buyer power (i.e., “monopsony” – see Manning 2013), or of “efficiency wages” (Leonard et al 2014) to explain the perceived market failure. Ironically, these two theories have diametrically opposed views of what happens in free markets. The monopsony theory implies that wages are too low: firms operate with unfilled vacancies, because raising wages enough to eliminate the backlog would require pay increases for all. But the efficiency wage theory implies that wages are in a sense “too high”: wages are above the market-clearing rate because paying well is a cheap way to help supervisors generate extra employee effort.
In fact neither theory fits well with the UK firm size structure which is shown in Table 4. We see that the UK has 3.6 million enterprises which only employ the owner, and obviously have no monopsony power or difficulty with supervision. There are also 1.2 million enterprises which employ workers, but the vast majority of these, 97%, employ fewer than 50 workers, and again can have no monopsony power or supervision issues. In fact, it is only the 6 thousand firms that employ more than 250 to which these theories might apply. However, these firms tend to pay higher wages in any case and so the minimum wage is broadly irrelevant for them. It is much more likely that unskilled and inexperienced workers lose from the minimum, and indeed their unemployment rises more than proportionately as the minimum rises, as shown in Gorry’s (2013) elegant model Monopsony and efficiency wage theories would seem irrelevant here.
International Evidence
It is fair to say that the UK evidence on the minimum wage does not find large employment effects. It does find some disturbing trends. However, the background to the UK’s minimum wage ensures that any statistical analysis is likely to lead to inconclusive results. For a more successful analysis, we need to look at countries where there is high variation of the minimum either for regions within a country, as in Canada or the USA, or at the variation provided by cross-country panels. A country with low welfare payments (e.g. the US) is also easier to analyse. Let us consider first Canadian and US results, and then examine South Africa which starkly underlines how minimum wages can be misused by skilled workers to cut out unskilled competition.[footnoteRef:3] [3: Another interesting country to consider could be France with its exceptionally high minimum (Gorry 2013). But France presents the same problems as the UK, that is, lack of within-country variation, plus the confounders of high welfare payments and high taxes. China is also a possibility, with good variation provided by different minimum wages in different cities. Fang and Lin (2013) provide evidence of strong negative minimum wage effects using city data which are better than the province data used by Wang and Gunderson (2012) who find inconclusive effects at least for Eastern China. However, China is too early to draw conclusions from yet, since it is so large and heterogeneous with complications of migrant labour and a large state-owned sector. There is also research on minimum wages in LDCs in general, bringing in non-compliance issues in the informal sector, which we will touch on below. ]
Canada provides one of the best conditions for research into the effects of minimum wages, since the ten Canadian provinces have different minimum wage policies, sometimes with considerable bite. Good time series data are also available. A convincing body of Canadian literature has thus built up, starting with Baker et al’s (1999) study of 9 provinces over 1975-1993. This study finds that a 10% increase in the minimum reduces teenage employment by 2.5%, and that it takes 6 years for the full effect to be revealed. Canadian data are also used in the recent study for 1997-2008 by Campolieti et al (2014, 587), who find a short-run elasticity of -0.16 for the 15-24 year group. Importantly, they note that their method cannot capture long-run minimum wage effects (since they follow individuals for only 6 months), and recommend doubling this elasticity to derive the full picture. This would lead to a long run elasticity of demand for labour of about -0.3, meaning that a 10% increase in the minimum wage would reduce employment by about 3% among the affected group. A similar finding for teen employment is made by Sen et al (2011). Interestingly, older workers’ employment appears to increase with minimum wage increases (Fang and Gunderson 2009), suggesting they are substituted for less productive youths, as we have already seen for the UK. The minimum’s adverse effect using good data thus becomes clearer.
A possible reason for the clarity of the Canadian effect is that Canada’s minimum wage workers tend more to be in the tradeable goods, exporting sectors. In this situation, a higher minimum quickly undermines competitiveness and causes employment reduction. The position is different if minimum wage workers are concentrated in non-tradeable activities such as retailing or construction (as in the UK). In this case, a rise in the minimum wage simply “takes wages out of competition” and this result can even be advantageous, especially for large firms (see Cox and Oaxaca 1982, more recently Neumark and Wascher 2008). Costs go up, but the increase is faced by everyone, and prices can increase to offset this given the absence of overseas competition. This factor might account for the weaker disemployment results found in the US studies (e.g. Addison et al 2012) of the restaurant sector noted below. Magruder’s (2013) study of minimum wages in Indonesia emphasises the importance of whether the sectors mainly affected are in the tradeable or non-tradeable sectors.
Turning to research on the US, there is now much technical controversy raised by the work of Allegretto et al (2011) and Dube et al (2010), well summarised in the recent work by Neumark et al (2014). The key problem is how to specify control groups, whether by allowing for state-specific trends, or by using contiguous states, which need not be good controls since cross-border counties might not experience the same shocks. Still, Neumark et al’s exhaustive analysis (2014, p627) concludes that, when the time trends are correctly specified, the elasticity of teen employment to the minimum wage remains in or near the -0.1 to -0.2 range. Elasticity of employment in the restaurant sector is lower, but still negative and significant at around -0.05 or -0.06 (2014, p644). Thus the adverse effect remains.
We cannot leave US minimum wage research without mentioning the famous but weak Card and Krueger (1995) studies of the response of fast-food restaurant employment to increases in minimum wages. The best-known of these studies is the contrast of New Jersey with Pennsylvania (which had no increase in the minimum wage). This research is the basis for stating that the conventional economic view that minimum wages cause unemployment is a “myth”. But the New Jersey versus Pennsylvania study is only based on 4 datapoints. The fact that there are many restaurants in the 4 samples (New Jersey before and after, and the control, Pennsylvania before and after) is no help since the same minimum wage regime applies in each. It is also worth noting that again we have the restaurant sector, which is sheltered from international competition, and so should have smaller employment reactions. The work is sold as “a powerful new challenge to the conventional view”, but it is misleading.
The evidence from South Africa shows what happens when minimum wages really go wrong. South Africa under white control, before 1994, had what it described as a “civilised labour policy” aiming to favour white employment (Van der Horst 1942, 250; also Siebert 1986). A pillar of this policy was high minimum wages and extended collective agreements which meant that only white workers, who were generally better educated than non-white workers, could gain employment. Minimum wages were thus used as a weapon against the majority. The higher costs that resulted were not so much of a problem when it came to employment because tariff barriers prevented imports competing with domestic businesses. The high minimum wage system continues to this day with extended collective agreements in particular supporting a strongly unionised African labour elite (Schultz 1998). For example, a union worker in manufacturing receives 70% more than a non-union worker (Schultz 1998, p700). The system has since been extended to agriculture and domestic service. In agriculture the pay increase has been large (Bhorat et all 2012), 17%, and employment has contracted considerably, by 14%. In domestic service, again protected from international competition, the effects might not have been so bad (Dinkelman 2012, Hertz 2005), but only 25% of households appear to comply. Thus, we see a policy originally designed to hurt African workers is now being carried forward by African politicians and unions themselves, and still hurting African workers.
Finally, let us consider the evidence from international cross-country panels. This research design gives the most variation in minimum wages, and thus helps create more robust studies. Negative employment effects from minimum wages are clear in all the studies. Admittedly there are some difficulties of comparability. In particular, countries such as Germany, Italy, Denmark and Sweden have no national minimum wage as such, but use legally enforceable extended collective agreements. Still, such agreements are effective minima (see below). There is also the difficulty of allowing for widely different welfare regimes, and these studies typically use the OECD’s index of gross benefit replacement rates which leaves out housing benefit that are important in the UK[footnoteRef:4]. However, hopefully the gross replacement rates capture the trends, and in any case the studies all control for country and time fixed effects. [4: Hence Addison and Ozturk’s (2012, Table 2b) replacement rate for the UK averages only 20%, much lower than the actual 60% shown in Table 1 above.]
Neumark and Wascher (2004) provide the first comprehensive treatment, analysing 17 OECD countries over the period 1975-2000. Their main finding (2004, p243) is that the minimum wage elasticity of teenage (15-19) employment is significantly negative, in the -0.2 to -0.4 range. More recently there have been OECD-based international studies by Dolton and Bondiabene (2012) of youth employment, and Addison and Ozturk (2012) of female employment. Dolton and Bondiabene (2012) also find a large negative effect of minimum wages for youth employment, with an elasticity of -0.3 to -0.4, most of this result coming in recession as might be expected. Addison and Ozturk (2012, Table 4) find a negative effect for the adult female employment-to-population rate, with an elasticity of -0.14 in the short run, and more in the long run when lagged effects are taken into account. Interestingly, there is an indication in this study (2012, Table 7) that the elasticity is larger (-0.34) in countries and time periods when employment protection law (EPL) is strict, as we would expect (see below). Thus, there is an unambiguous picture of strong negative minimum wage effects on lower-productivity groups in the international panel-based literature, which provides arguably the best foundation for research.
Collectively set minimum wages
Collectively set minimum wages arise when a collective agreement is extended by law to third parties within an industry or sector. A detailed set of minimum wages (and conditions) covering many job types and levels is thereby established for the industry or sector. Martins (2014) shows how the process works in the case of Portugal. Such extensions are the result of so-called “erga omnes” (towards others) regulations, and have the aim of reducing non-union low-wage competition. They are common (see Murtin et al 2014) in countries where the unions are politically powerful – for example, are part-funded or privileged by the state as in Greece or France or South Africa – but where local union power is low (again, France and Greece[footnoteRef:5]). In these circumstances, unions have the power to bring about these regulations and also need to, since non-union firms are so prevalent. The picture is given in Table 5, where we see that France, Spain, Portugal, Germany and Italy all have high use of extensions arrangements. The important point is that erga omnes regulations enable the setting of detailed minimum wage floors, floors that are determined by big business and labour in the capital cities – Athens, Rome, Paris – with little concern for conditions in the provinces. Hence, a straitjacket of minimum wages is thrown over the country. [5: The UK used to have an erga omnes arrangement for unions to petition for extension of their agreements, but Thatcher dismantled it (see Addison and Siebert 2000) with the 1980 Employment Act. The “fair wages resolution” requiring government contractors to observe terms no less favourable than those obtaining under collective agreements was also dropped at this time.]
Research on this type of minimum wage setting builds on the literature of union power raising unemployment, which of course has long been controversial because of the many factors that influence unemployment. Steven Nickell’s (1997) famous study of unemployment in OECD countries in the 1980s and 1990s shows union agreement coverage strongly raising unemployment, but his “bargaining coordination” variable washes out this effect, suggesting that cooperation between the two sides of industry (corporatism) could engender responsible unionism. Recent work on these lines is reported in the OECD’s (2011b, p152) study of inequality using the OECD panel of countries over 1984-2007. Here there is shown to be a well-determined negative effect of collective bargaining coverage on employment rates. High tax rates and employment protection legislation hurt too. The most recent work using the OECD panel is by Murtin et al (2014) with a more extensive model, this time of unemployment. Their main innovation is to use collective agreement coverage extension, as shown in Table 5, which they find interacts with taxes (also shown in Table 5) to raise unemployment. Their minimum wage variable also raises unemployment, again more so when taxes are high.
There are also instructive results from two country case studies of Portugal and Greece involving extended agreements. As can be seen from Table 5, both of these countries have used such agreements extensively, though they are now restricted by recent debt bailout agreements[footnoteRef:6]. Martins’ (2014) analysis of Portugal over 2007-11 links unemployment to extensions of agreements to non-parties, and shows that average employment levels in affected sectors drop by 2% in the four months following extensions, as firms stop hiring, and close down. Since the wage increase is 2% to 4%, the implied elasticity is between -0.5 and -1 (Martins 2014, 14). Peripheral employment of temporary workers and sub-contractors meanwhile increases, as we would expect. This temporary worker result is the same as that for the Greek study (Anagnostopoulos and Siebert 2013), based on a survey of 200 provincial firms, which finds that low-paying firms, near the minimum, are more likely to employ temporary workers. The Greek study also shows that these effects persist even though many Greek firms do not in fact pay the minimum they remain small so as to avoid the attentions of the labour inspectorate. The minimum wage and extended collective agreements coupled with high taxes thus not only reduce employment, but also push firms to be too small[footnoteRef:7]. [6: In Greece for example (see Commission 2014 p49 and LABREF database), the government in 2011 and 2012 agreed to reduce minimum pay rates by 22% (32% for young people). It also agreed to suspend extension of occupational and sector collective agreements, and to allow firm-level agreements which could be less favourable than the sector-level agreement. According to IndustriALL (2014) measures such as this caused Greek sector agreements to fall from 65 to 14 in 2013, and in Portugal (Martins 2014), coverage fell from 1.8 million in 2008 to 290,000 in 2012.] [7: Thus in Greece and Portugal firms are too small, with only 5% employing more than 10 people, compared to an OECD average of 15% employing more than 10 (see OECD, 2011c). Alternatively, if policing of the extended agreements is effective, as in South Africa, small firms can be prevented from growing enough (see Magruder 2012). The point is, regulation breeds regulation.]
Conclusions and thoughts on real help for the unskilled
Mnimum wages do have adverse employment effects broadly in line with conventional economic models. UK studies generally find small effects, but the UK environment is poorly suited to measure such effects given the changes in welfare benefits, and the lack of regional variation. Matters become clearer when we turn to more suitable empirical settings which provide clear evidence of an adverse effects of minimum wages on employment.
Better ways to help the poor involve raising skills or, in the shorter term before such policies take meaningful effect, the provision of wage subsidies through the welfare system. When it comes to raising skills, the family is crucial, together with education. In fact, Table 6 shows that comparable countries all have difficulties with families in poor circumstances, so the UK is not alone. Admittedly, the first two rows show that the UK has exceptionally high numbers of children in single parent and workless households (see Kristian Niemietz 2012). Yet while Italy does best on this criterion, as can be seen, it lags otherwise. The intergenerational earnings correlation is 0.48 in the UK which is high[footnoteRef:8], but no worse than in Italy or the US. Moreover, inequality in disposable household incomes, as measured by the Gini coefficient, is broadly similar at around 0.30 to 0.34 in all the countries. Only the US stands out as particularly unequal, but this result is to be expected since the US is much larger and more heterogeneous. [8: Here we correlate the incomes of fathers and sons. A low correlation as in Germany, 0.32, is better because it shows that good parental circumstances have less effect in improving the child’s life chances than in the UK which has a much higher correlation.]
Furthermore, taking adult language proficiency which is an important measure of skill, we see a distinct gradient according to family background, as is to be expected. Individuals from advantaged backgrounds (having at least one parent with a university degree) score about 1 point higher on a 1 to 5 scale than the less advantaged. All the comparator countries have an almost exactly similar effect, which points once again to the fact that this problem is deep-seated. Encouragement of stable families and better, less unionised and more competitive school systems is difficult, but they are important policy priorities. High welfare simply makes matters worse in the long term, as well described by Charles Murray (1984, 2012).
A quick and well-targeted way of helping the working poor is by subsidising low paid work, a policy which contradicts the minimum wage. People have their earnings topped up by the tax payer, and are therefore prepared to work for less, which expands their job opportunities. In fact, such a policy has been in place in the UK ever since 1999 when the Working Families Tax Credit was introduced (Azhmat 2006), modelled on the US Earned Income Tax Credit, and since expanded with Working Tax Credit (see Bourne and Shackleton 2014). Figure 4 shows how the policy works by reducing the wage that the employer pays from W* to WL. The equilibrium on the original demand curve moves to point A, to which is added the subsidy, and the worker receives the higher gross wage, WH.= WL+SUB. It is important that the wage the employer pays is allowed to fall, though, and if the minimum wage prevents this fall, the policy will not work, and we remain at C. In a sense, the employer has to be paid some of the subsidy and his/her profits will increase which has disturbed some commentators (e.g. Gregg et al 2012). However, the result is more jobs which is what we want.
Tax credits conditional on work currently encourage about 2.5 million workers to work (see Browne and Hood 2012), but they only account for about 5% of the amount paid out to working age welfare recipients, and are dwarfed in particular by child tax credits and housing benefits. Reform is needed to make working tax credits more important (see Bourne and Shackleton 2014). This said, the way they lower wages (see Azmat 2006 and Hotz and Scholz 2000) is exactly as predicted by our conventional labour market analysis.
In conclusion, there is no need for analysts (e.g. Holmlund 2014, or Schmitt 2013) to worry about “discernible effects" of minimum wages on employment. The effects are discernible, when properly measured. Hence it is indeed probable that the UK’s dismal youth labour market performance since 1999 is partly attributable to the imposition of the minimum wage – interacting with high tax rates. This is also the case with the poorly functioning youth labour markets of Portugal and Greece, and others such as France and South Africa. Moreover, the way in which more skilled workers displace the less skilled, and temporary workers displace permanent workers is also in line with conventional economic models. Obviously, in a political world which denies productivity differences – including skills, gender, age and disability differences[footnoteRef:9] – the differential effects of minimum wages are politically unwelcome This is all the more reason for economists to stick to their guns and look for real, not fake, ways to help the poor. [9: The disability pressure groups often do the disabled no favours. For example, (LPC 2005 p124), Mothercare for Children in Hospital Ltd (MCCH) cooed: “ the minimum wage is a positive step to reducing stigma, discrimination and workplace exploitation…it has acted as a catalyst to change” etc etc. But Shaw Employment Services is blunt: ”Both client and provider are finding that instead of being more able to help the disabled achieve employment, the Government, through the NMW, has inadvertently created the first barrier” (LPC 2003, p105).]
References
Addison, J., M. Blackburn and C. Cotti (2012), “Labour Market Outcomes: County-Level Estimates from the Restaurant-and-Bar Sector”, British Journal of Industrial Relations, 50: 412-435.
Addison, J. and O. Ozturk (2012), “Minimum Wages, Labor Market Institutions, and Female Employment: A Cross-Country Analysis”, Industrial and Labor Relations Review, 65: 779-809.
Addison, J. and W. Stanley Siebert (1994), “Recent Developments in Social Policy in the New European Union”, Industrial and Labor Relations Review, 48:5-27.
Addison J and W. S. Siebert (2000), "Labor Market Reform in the UK: from Thatcher to Blair", Journal of Private Enterprise, 15: 1-34.
Allegretto, S., A. Dube, and M. Reich. (2011), “Do Minimum Wages Really Reduce Teen Employment? Accounting for heterogeneity and selectivity in state panel data. Industrial Relations 50: 205–40.
Azmat, G. (2006), “The Incidence of an Earned Income Tax Credit: Evaluating the Impact on Wages in the UK”, London School of Economics, CEP Discussion Paper 724.
Anagnostopoulos A. and W. S. Siebert (2012), “The impact of Greek labour market regulation on temporary and family employment - Evidence from a new survey”, IZA Discussion Paper 6504.
Baker, M., D. Benjamin and S. Stanger (1999). “The Highs and Lows of the Minimum Wage Effect: A Time-Series Cross-Section Study of the Canadian Law”, Journal of Labor Economics, 17:18-50.
Baum, C. (2006), An Introduction to Modern Econometrics Using Stata, College Station, Texas: Stata Press.
Bhorat, H., R. Kanbur and B. Stanwix (2013), “Estimating the Impact of Minimum Wages on Employment, Wages and Nonwage Benefits: The Case of Agriculture in South Africa”, Cornell University, School of Applied Economics and Management, WP 2013-05.
Bourne, R. and J. Shackleton (2014), “The Minimum Wage: Silver Bullet or Poisoned Chalice”, Institute of Economic Affairs Briefing 14.01.
Brown J. and A. Hood (2012), A Survey of the UK Benefit System, IFS Briefing Note BN13, London: Institute for Fiscal Studies.
Campolieti, M., M. Gunderson and B. Lee (2014), “Minimum Wage Effects On Permanent Versus Temporary Minimum Wage Employment”, Contemporary Economic Policy, 32: 578-591.
Card, D., and A. Krueger (1995). Myth and Measurement: the New Economics of the Minimum wage, Princeton: Princeton University Press.
Commission (2006), Green Paper—Modernising labour law to meet the challenges of the 21st century, Brussels 22.11.2006 COM(2006) 708 final.
Commission (2014), “The Second Economic Adjustment Programme for Greece”, Occasional Papers, No. 192, April, Brussels: Directorate-General for Economic and Financial Affairs .
Corak, M., (2013), Income Inequality, Equality of Opportunity, and Intergenerational Mobility, Journal of Economic Perspectives, 27: 79-102.
Cox, J. and R. Oaxaca (1982), ‘The political economy of minimum wage legislation”, Economic Inquiry, 20: 533-50.
Dickens, R., R. Riley and D. Wilkinson, (2012), “Re-examining the Impact of the National Minimum Wage on Earnings, Employment and Hours: the Importance of Recession and Firm Size”, University of Sussex and National Institute of Economic and Social Research. http://www.sheffield.ac.uk/polopoly_fs/1.247213!/file/E1_riley.pdf
Dinkelman T. and V. Ranchod (2012), “Evidence on the impact of minimum wage laws in an informal sector: Domestic workers in South Africa”, Journal of Development Economics, 99: 27-45.
Dolton, Peter and Chiara Bondibene (2012), “The International Experience of Minimum Wages in an Economic Downturn”, Economic Policy, 27: 99-142.
Dube, A., T. Lester, and M. Reich. (2010), Minimum Wage Effects Across State Borders: Estimates using contiguous counties. Review of Economics and Statistics 92: 945–64.
Fang, T. and M. Gunderson (2009), “Minimum Wage Impacts on Older Workers: Longitudinal Estimates from Canada”, British Journal of Industrial Relations, 47: 371-87.
Fang T., and C. Lin (2013), “Minimum Wages and Employment in China”, IZA Discussion Paper 7813.
Godfrey, S., J. Theron and M. Visser (2007), The State of Collective Bargaining in South Africa, UCT: Development Policy Research Unit Research Paper 07/130
Gorry, A. (2013), “Minimum wages and youth unemployment”, European Economic Review 64: 57-75.
Gregg, P., A. Hurrell and M. Whittaker (2012), Creditworthy – Assessing the Impact of Tax Credits, London: Resolution Foundation.
Hertz, T. (2005), “The Effect of Minimum Wages on the Employment and Earnings of South Africa's Domestic Service Workers, Upjohn Institute Working Paper 05-120.
Holmlund, B. (2014), “What do labor market institutions do?”, Labour Economics 30: 62-69.
Hotz, V. and J. Scholtz (2000), “Not Perfect, but Still Pretty Good: The EITC and Other Policies to Support the US Low-Wage Labour Market”, OECD Economic Studies 31.
IndustriALL (2014), “Negotiating our future!”, Collective Bargaining and Social Policy Conference, Vienna, http://www.industriall-europe.eu/committees/CB/2014/Increasing%20cover%20rate-EN.pdf
Kay L. (2010), Escaping the Poverty Trap, London: Policy Exchange.
Krugman, P. (1994), “Past and Prospective Causes of High Unemployment”, Federal Reserve Bank of Kansas City, http://www.kc.frb.org/publicat/econrev/pdf/4q94krug.pdf
Leonard, M., T. Stanley and H. Doucouliagos (2014), “Does the UK MinimumWage Reduce Employment? A Meta-Regression Analysis”, British Journal of Industrial Relations, 52: 499-520.
Living Wage Commission (2014), Work That Pays, London: Joseph Rowntree Trust.
LPC (2014), Low Pay Commission Report 2014, Cm 8816, London: Low Pay Commission.
LPC (2005), Low Pay Commission Report 2005, Cm 6475, London: Low Pay Commission.
LPC (2003), Low Pay Commission Report 2003 – Building on Success, London: Low Pay Commission.
LPC (2001), The National Minimum Wage – Making a Difference, Volume 1, Cm 5075, London: Low Pay Commission.
Magruder, J (2013), “Can Minimum Wages Cause a Big Push? Evidence from Indonesia”, Journal of Development Economics, 100: 48-62.
Magruder (2012), High Unemployment Yet Few Small Firms: The Role of Centralized Bargaining in South Africa”, American Economic Journal: Applied Economics, 4: 138-166.
Manning A. (2013), “Minimum Wages: A View from the UK”, Perspektiven der Wirtschaftspolitik, 14: 57–66.
Martins, P. (2014) “30,000 Minimum Wages: The Economic Effects of Collective Bargaining Extensions”, IZA Discussion Paper 8540.
Murray C. (1984), Losing Ground: American Social Policy 1950-1980, New York: Basic Books.
Murray C. (2012), Coming Apart: The State of Whate America 1960-2010, New York: Crown Forum.
Murtin, F., A. de Serres and A. Hijzen (2014), “Unemployment and the coverage extension of collective wag eagreements”, European Economic Review, 71: 52-66.
Neumark, D., J. Salas and W. Wascher (2014), “Revisiting the Minimum Wage-Employment Debate: Throwing Out the Baby with the Bathwater?”, Industrial and Labor Relations Review, 67 (supplement): 608-48.
Neumark, D. and W. Wascher (2008), “Minimum Wages and Low-Wage Workers: How Well Does Reality Match the Rhetoric?”, Minnesota Law Review, 92: 1296-1316.
Neumark, D., and W.Wascher, 2004. “Minimum Wages, Labor Market Institutions and Youth Unemployment: A Cross-National Analysis”, Industrial and Labor Relations Review, 57: 223-48.
Nickell, S. (1997), “Unemployment and Labor Market Rigidities: Europe versus North America”, Journal of Economic Perspectives, 11: 55-74.
Riley, R. (2013), “Modelling Demand for Low Skilled/Low Paid Labour: Exploring The Employment Trade-Offs of a Living Wage”, NIESR Discussion Paper 404
Niemietz, K. (2012), Redefining the Poverty Debate, London: Institute of Economic Affairs.
OECD (2014), Education at a Glance 2014: OECD Indicators, OECD Publishing, http://dx.doi.org/10.1787/eag-2014-en
OECD (2011), Doing Better for Families Paris: OECD Publishing DOI:10.1787/9789264098732-en
OECD (2011), Divided We Stand: Why Inequality Keeps Rising, Paris: OECD Publishing. http://dx.doi.org/10.1787/9789264119536-en
Sen, A., K. Rybzynski and C. Van De Waal (2011), “Teen Employment, Poverty, and the Minimum Wage: Evidence from Canada”, Labour Economics, 18: 36-47
Schmitt, J. (2013). “Why Does the Minimum Wage Have No Discernible Effect on Employment?”, Washington: Center for Economic and Policy Research
Schultz, T. and G. Mwabu (1998), “Labor Unions and the Distribution of Wages and Employment in South Africa”, Industrial and Labor Relations Review, 51: 680-703.
Siebert W. S. (2014), “The Living Wage”, Economic Affairs Magazine, Summer: 18-21.
Siebert W. S. (2013), “The Wage Flaw – Why Minimum Pay Isn’t Working”, Economic Affairs Magazine, Summer: 4-7.
Siebert, W. S. (1986), “Restrictive Practices in South Africa’s Labour Market”, Economic Affairs, October-November: 26-29.
Van der Horst, S. (1942), Native Labour in South Africa, London: Frank Cass.
Visser J. (2013), Wage Bargaining Institutions - from crisis to crisis, Brussels, Directorate-General for Economic and Financial Affairs: Economic Papers 488.
Wang, J. and M. Gunderson (2012), “Minimum wage effects on employment and wages: dif-in-dif estimates from eastern China”, International Journal of Manpower, 33: 860-876.
Table 1: Net replacement rates over 60 Months unemployment
|
|
France |
Germany |
United States |
UK |
||||
|
|
No child |
2 child |
No child |
2 child |
No child |
2 child |
No child |
2 child |
|
2001 |
60.3 |
70.3 |
59.6 |
73.5 |
15.2 |
43.0 |
55.0 |
64.9 |
|
2012 |
53.6 |
63.2 |
49.9 |
70.4 |
24.8 |
42.8 |
51.0 |
72.4 |
|
Change 2012-2001 |
-6.7 |
-7.1 |
-9.7 |
-3.0 |
9.6 |
-0.2 |
-4.0 |
7.5 |
Source: OECD Benefits and Wages: Statisticshttp://www.oecd.org/els/benefitsandwagesstatistics.htm (NRR_Over5years_EN)
Notes: Replacement rates are calculated based on incomes after any tax and social security contributions have been deducted, and any cash benefits received. Family qualifies for cash housing assistance and social assistance "top ups". The figures give unweighted averages relative to full-time earnings levels of 67% and 100% of average worker earnings. (Prior to 2001 only gross rates not including housing assistance are available.)
Table 2: Adverse Changes in Employment for Unskilled Workers
Labourforce participation rate (% of the population in each group)
|
|
All working age |
No qualific-ations |
Disabled people |
18-20 yrs old |
16-24 yrs old |
France 15-24 |
|
2000 |
71.7 |
50.8 |
37.7 |
61.2 |
69.7 |
35.6 |
|
2013 |
71.6 |
42.6 |
42.1 |
47.0 |
61.7 |
37.6 |
|
Change 2013-2000 |
-0.1 |
-8.2 |
4.4 |
-14.2 |
-8.0 |
2.0 |
Sources: Commission (2014, Table 2.11), and OECD Employment Outlook (2014)
Table 3: NEETs – Country Trends
(% of 15-19 year olds not in education, employment or training)
|
|
France |
Germany |
Italy |
United States |
UK |
OECD average |
|
1997 |
2.9 |
5.0 |
15.2 |
7.1 |
8.0 |
9.2 |
|
2012 |
6.9 |
3.0 |
12.0 |
7.7 |
9.5 |
7.6 |
|
Change 2012-1997 |
4.0 |
-2.0 |
-3.2 |
0.6 |
1.5 |
-1.6 |
Sources: OECD Education at a Glance (2014), Chart C5.3, and OECD Doing Better for Families (2011), Figure 1.15
Table 4: Firm Size and Employment, UK 2009
|
Number - Private Sector, 2009 (including pub corps and nationalised bodies) |
|||
|
|
Enterprises (000) |
Employees (mill.) |
|
|
Enterprises. without employees |
3620 |
0 |
|
|
Enterprises with employees |
1220 |
18.2 |
|
|
Enterprise sizes: |
1-4 |
795 |
1.8 |
|
|
5-49 |
390 |
4.6 |
|
|
50-249 |
30 |
2.6 |
|
|
250-499 |
3 |
1.0 |
|
|
500 or more |
3 |
8.1 |
|
MEMO |
|
|
|
|
Central and local govt |
5 |
5.4 |
|
|
Non-profit |
84 |
1.9 |
Source: BIS, Enterprise Directorate, 2010, http://stats.bis.gov.uk/ed/sme/SMEStats2009.xls#'UK Whole Economy'!A1
Notes: An enterprise is the smallest group of legal units which has autonomy. It is based on the Interdepartmental Business Register (IDBR) formed from VAT or PAYE records collected by HMRC. Since the VAT threshold (£67k in 09) excludes small firms, estimates of their numbers are then added using Labourforce Survey figures of numbers of self-employed (4.1m). Private households and temp agencies are excluded..
There is no lower bound for inclusion as an enterprise, hence the smallest amount of enterprise activity counts – hence there are many "no employee" enterprises which have only working proprietors in the business. “Employees” have a contract of employment, and include part-timers. Working proprietors are self employed (but working directors of companies are counted as employees).
Table 5: Collectively Set Minimum Wages, early 2000s
|
|
Can |
Den |
Fra |
Ger |
Gre |
Ita |
Nor |
Por |
SA |
Spain |
Swe |
UK |
US |
|
Coverage* extension |
4 |
0 |
79 |
48 |
30 |
46 |
15 |
49 |
10 |
57 |
1 |
9 |
4 |
|
Use of erga omnes clauses |
No |
No |
High |
High |
High |
High |
Some |
High |
Some |
High |
No |
No |
No |
|
Union density |
31 |
75 |
9 |
28 |
30 |
34 |
56 |
25 |
25 |
15 |
81 |
34 |
14 |
|
Govt rev % GDP |
40 |
55 |
50 |
45 |
38 |
47 |
57 |
40 |
27 |
36 |
54 |
40 |
31 |
Sources: Murtins et al (2014), Industriall Global Union (2014) http://www.industriall-europe.eu/committees/CB/2014/Increasing%20cover%20rate-EN.pdf; Visser (2013), Godfrey (2007), OECD Government at a Glance (2013, Table 3.11.
Notes: * coverage extension measured as collective agreement coverage minus percentage unionisation
Table 6: Inequalities – Country Comparisons
|
|
France |
Germany |
Italy |
United States |
UK |
|
|
% children (0-18) in single-parent households, 2007a |
14 |
15 |
10 |
22 |
26 |
|
|
% children (0-14) living in jobless households, 2011b |
10 |
7 |
8 |
8 |
19 |
|
|
Intergenerational earnings correlation, late 90sc |
0.40 |
0.32 |
0.49 |
0.50 |
0.48 |
|
|
Gini coefficient of inequality of household disposable incomed |
0.29 |
0.30 |
0.33 |
0.37 |
0.34 |
|
|
Average adult literacy proficiency levele |
Both parents<upper secondary educ. |
2.0 |
1.8 |
1.9 |
1.8 |
2.1 |
|
|
At least one parent with tertiary educ. |
3.0 |
2.8 |
2.7 |
2.9 |
3.1 |
Sources: a OECD Doing Better for Families (2011) Table 1.1; b OECD Doing Better for Families (2011) Table LMF1.1.A; c Corak (2013, p82); d OECD Divided We Stand (2011), Fig 6.1. (income is household income, corrected for household size – the Gini coefficient varies between 0 for perfect equality, and 1 for perfect inequality); e OECD Education at a Glance (2014, Chart A4.4) (literacy score is from the OECD programme for International Assessment of Adult Competencies, with proficiency graded from worst (1) to best (5)).
( D S Welfare benefits floor W* A B Unskilled Labour E Minimum wage floor )
Figure 1: Welfare benefits confound minimum wage employment effects
Figure 2: Effects of the minimum wage on earnings - 1998 and 1999 compared
Source: Low Pay Commission Report 2000, p19
( Unskilled Labour Minimum wage or collective agreement floor D| gross S W* E 2 B E 1 D| net tax TAX )
Figure 3: Minimum wage magnifies disemployment effects of tax
( Unskilled Labour W* SUB D| gross D| plus subsidy S W* W L W H B A Minimum wage or collective agreement floor C )
Figure 4: Subsidising work
5
