1
Thesis BSc. Economics and Business
Specialisation: Economics
University of Amsterdam
The Differences in Minimum Wage Regulations as
Determinant of Youth Employment and Youth
Unemployment:
A Cross-Sectional Analysis
Name Menouschka Claudine Plugge
Student number 10364757
Field Labour Economics
Supervisor Gabriele Ciminelli
Date 02/02/2016
Abstract
In this study, the differences in minimum wage regulations as a determinant of youth employment and youth unemployment was examined. The relationship between the differences in minimum wage regulations and the youth employment has been examined on the macroeconomic level using a cross sectional analysis between the 28 European Union countries in three different time periods. The study shows positive significant effects on the relationship between the differentiation of an age-dependent minimum wage and the youth employment.
2
Table of contents
1. Introduction . . . . . . . . 3
2. Literature review . . . . . . . . 4
2.1. Theory . . . . . . . . 4
2.2. Previous empirical evidence . . . . . 6
3. Descriptive Statistics . . . . . . . 8
3.1. Minimum wages and youth employment and youth unemployment in the European Union . . . . . 8
4. Methodology and Data . . . . . . . 12
4.1. Methodology . . . . . . . 12
4.2. Data . . . . . . . 15
5. Results . . . . . . . . . 15
6. Conclusion and Discussion . . . . . . 18
6.1. Conclusion . . . . . . . 18
6.2. Discussion . . . . . . . 21
7. Literature . . . . . . . . 22
Statement of Originality
This document is written by Menouschka Plugge who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
1. Introduction
The United Nations for statistical purposes defines youth as those persons in the age group 15 to 24 (2008). This group is further divided in teenagers aged 15 to 19 and young adults aged 20 to 24. In 2014 youth unemployment in the European Union was 22.2%, this was more than double the unemployment rate of 9.0% among adults aged 25 to 74 (Eurostat). Youth unemployment remains high compared to other age groups in most member states of the European Union. Thus youth unemployment is a policy relevant problem in the European Union. This thesis aims to investigate the minimum wage as an important determinant of youth employment and unemployment.
Youth unemployment is of particular burden to the group of unemployed youth. Demoralization is a problem for people who become unemployment during their early years on the labour market, they may lose the confidence or hope in finding a job. Moreover, people who get unemployed after leaving tertiary education may see a depreciation in their human capital and a decline in their employment prospects, which again may lead to demoralized thinking. These discouragements of getting employed during their early working years could lead to social exclusion (Gomez-Salvador and Leiner-Killinger, 2008, p. 8).
Meanwhile, youth unemployment is not only a concern for those who are affected by it, but also for the total economy of a country. First, unemployment among youth implies that labour is not utilized to its full potential. This has a negative effect on the potential growth of the economy. Second, youth unemployment implies that there is less labour input of those who are supposed to innovate the production developments with their newly gained expertise and more up-to-date required knowledge(Gomez-Salvador and Leiner-Killinger, 2008, p. 8).
In general, there are three major causes of high youth unemployment. First, youth employment and youth unemployment are highly sensitive to the strength of the economy as a whole. Periods with high aggregate level of economic activity and high level of adult employment will result in a period with high youth employment. This implies that if adult unemployment is high, youth unemployment will also be high. Second, the determent of youth employment is correspondent to the proportion of youth in the population. The proportion of youth in the population is negatively related to youth employment. If the proportion of youth in the population is high, youth employment will be low and vice versa. Thus, the relative proportion of youth in the population is an important determent of youth employment and youth unemployment. Therefore, the emphasis in this study is on the relative position of youth employment to total employment and on the relative position of
4
youth unemployment to total unemployment. Third, the minimum wage is an important determinant of youth employment (Freeman & Wise, 1982, p. 3).
Earlier mentioned, this study aims to investigate the minimum wage as an important determinant of youth employment and unemployment. The differences in minimum wage regulation for the member states of the European Union has been examined. Each state has its own regulations for the establishment of the minimum wage. The different minimum wage regulations within the European Union can be classified in three groups: countries without a national minimum wage, countries with a national minimum wage, and countries with a differentiated national minimum wage that is age-dependent. This study has examined the influence of the three different minimum wage regulations on the relative youth employment, relative youth unemployment and relative youth labour supply with regard to the total population. This influence has been examined on the macroeconomic level using a cross sectional analysis between the 28 European Union countries in three different time periods. Analysis on this data show positive significant effects on the relationship between the differentiation of an age-dependent minimum wage and the youth employment. This study consists of 6 sections. Section 2 deals with previous theoretical and empirical literature. Section 3 shows the descriptive statistics of minimum wages and relative youth employment, relative youth unemployment, and relative youth labour supply in the European Union. Section 4 describes the methodology and data. In Section 5 the results are presented and Section 6 consists of the conclusion and discussion.
2. Literature review
2.1 Theory
To clarify why minimum wage regulation is one of the major causes of youth unemployment, the classical theory about unemployment will be explained. The classical theory about unemployment is wage rigidity. Wage rigidity is the failure of wages to adjust to a level of equilibrium, the level at which labour supply equals labour demand. If the real-wage is above the equilibrium level, then the supply of labour exceeds the demand of labour. This implies that the quantity of labour supplied exceeds the quantity of labour demanded. The result is unemployment. The specialist term of this specific kind of unemployment is structural unemployment. To summarize, structural unemployment arises if there is a fundamental imbalance between the number of people who want to work and the number of jobs that are
5
available. There are three major causes for structural unemployment: (i) minimum wage laws, (ii) the power of unions and collective bargaining, (iii) the efficiency wages (Mankiw, 2013, p. 181-182). In this research the influence of different minimum wage regulations on relative youth employment and relative youth unemployment is further examined, this is why the other two causes of structural unemployment are omitted in this paper.
A country that implements a statutory minimum wage sets a legal minimum on the wages that firms pay their employees. The government causes wage rigidity by preventing wages from falling to the equilibrium level (Mankiw, 2013, p. 182-183). Setting a minimum wage above the equilibrium wage in competitive labour markets will result in an increase of the average and marginal cost of labour. Ceteris paribus, the increase of the average and marginal cost will cause firms to reduce the quantity of labour demanded. Thus resulting in unemployment (Rebitzer and Taylor, 1995, p. 245).
For most of the workers, this minimum wage is not binding, because they earn well above the minimum. Meanwhile for some workers, especially the unskilled and inexperienced, the minimum wage raises their wage above its equilibrium level (Mankiw, 2013, p. 182-183). This could prevent employers from hiring the workers who will earn less than the minimum wage in an equilibrium situation, resulting in structural unemployment. Thus the minimum wage excludes the lowest quality labour of the market. The lowest quality of labour are often young, unskilled or inexperienced workers (van Soest, 1989, p. 280).
Chart 1: Demand and Supply framework (Mankiw, 2013, p. 181)
In chart 1 the demand and supply framework of labour with a minimum-wage law is illustrated. In chart 1 the real minimum wage is illustrated while the government sets the
6
nominal minimum wage. Here the minimum wage sets a legal minimum on the wages that firms pay their employees. It shows that the quantity of labour demanded, at the minimum wage, is smaller than the quantity of labour supplied. This results in structural unemployment. After all, the costs of labour will increase and may result in higher costs relatively to the productivity of these lowest quality labour (Mankiw, 2013, p. 181-182). Thus, the classical theory suggests that implementing a national minimum wage will result in negative employment and unemployment effects.
Elimination of this lowest quality of labour from the market will potentially alters the entire wage distribution (Brown, Gilroy and Kohen, 1982, pp. 19-20). Moreover, some of the unemployed may think that the chances of finding a job are too weak to make actively searching for a job useful. People who are not actively searching for a job are not included in the official unemployment count. The division between unemployment and not-in-the labour force is therefore ambiguous (Brown, Gilroy and Kohen, 1982, pp. 19-20). This is why the impact of a national minimum wage can be underestimated by only considering the youth employment, youth unemployment, and youth labour supply.
The minimum wage effects are larger for the group of workers who would have earned less than the minimum wage in absence of the minimum wage regulation. This is, according to the supply and demand framework, the lowest quality labour of the market. In general, teenagers and youth are overrepresented in the group with the lowest quality labour of the market, considering that they are the least experienced workers in the labour supply. This is why studies of the minimum wage effect on teenagers and youth find greater effects than those of other population groups (Brown, Gilroy, and Kohen, 1982, pp. 47-49).
To eliminate the greater effects of the minimum wage on youth some countries implement lower minimum wage rates for younger and less experienced workers, these are the age-dependent minimum wages. Lower minimum wages for young workers is justifiable considering that the minimum wage effects are substantially larger for young workers than for adults (van Soest, 1994, p. 100). Implementing an age-dependent minimum wage may create an opportunity for low-productive youth to gain working experience and thus improve their position in the labour market. The age-dependent minimum wage may also encourages youth to pursue education. A low wage makes education more appealing (Vuuren and Bosch, 2012, p. 2).
7
There has been much research done on the effect of minimum wage on youth employment and youth unemployment. Even though theory suggest that a minimum wage has negative youth employment and youth unemployment effects, previous research has not been able to come to a consensus and finds ambiguous effects. This section summarizes the most important findings of minimum wage laws on youth employment and youth unemployment.
A study of 17 OECD countries over the period 1975-2000 find that increases in minimum wages tend to lead to employment losses among the youth. However, the study find that the negative employment effects will be smaller in countries that have implemented age-dependent minimum wage rates for youth. This article implies that an age-age-dependent minimum wage softens the negative employment effects for the youth arising from the minimum wage. It reverses some of the negative employment effects for the youth (Neumark and Wascher, 2004, p. 226-243).
In the US, an empirical study on a 10 percent increase of the minimum wage has been done. The 10 percent increase has estimated to result in about a 1-3 percent reduction in total teenage employment. The unemployment effects would be larger for those whose wages would otherwise be the lowest, young teenagers. The research also concludes that with an increase of the minimum wage adult employment will improve. The minimum wage protects adults from teenage competition. The results imply that implementing a national minimum wage will have negative employment effects for youth. It may suggest that implementing an age-dependent minimum wage will softens the negative employment effects for youth (Brown, Gilroy and Kohen, 1982, p. 32-42).
New Zealand has been through a large minimum wage reform affecting youth workers since 2001. Before the reform the statutory age for the adult minimum wage was 20. The age-dependent group ranged from 16 to 19 and had a youth minimum wage of 60 percent of the adult minimum wage. After the reform the statutory age for the adult minimum wage was lowered till 18. The age-dependent group ranges from 16 to 17 and has a higher youth minimum wage of 80 percent of the adult minimum. An empirical study examined the effects of this large reform in the minimum wages and the affecting youth workers in New Zealand. It found no conflicting effects on youth employment immediately following the reform in 2001. The study found some weak evidence of employment loss by 2003. This was about 2 percent for the age group 18-19, and 2-4 percent for the age group 16-17. It found a decline in educational enrolment and an increase in unemployment. This suggests that a more differentiated age-dependent minimum wage will result in smaller negative employment effects for youth. The age-dependent minimum wage will soften the negative employment
8
effects. Problem remains that the evidence is weak, the results stay ambiguous (Hyslop and Stillman, 2005, pp. 202-227).
Even though the theory suggest that a minimum wage has negative employment and unemployment effects for youth, there is some economic research that contradicts this theory. In the United States, empirical research relying on quasi-experimental evaluations of cross-sectional and longitudinal data has failed to find negative employment effects for young or low-wage workers with the implementation of a minimum wage (Hyslop and Stillman, 2005, p. 202). One of these quasi-experimental evaluations was on a case study of California by David Card. He examined the minimum wage effect on teenage employment in California. In July 1988, the minimum wage increased from $3.35 to $4.25 per hour and in the previous year 11% of all workers and 50% of California teenagers earned between this range of dollars per hour (Card, 1982, p. 52). Previous study suggest that teenage employment would decline by 3-8% in response to the rise in the minimum wage (Brown, Gilroy and Kohen, 1982, p. 32-42). Contradictory to this previous research, there was no significant decline in teenage employment. The experimental study did not find empirical support for the conventional prediction that previous economist make regarding the employment of minimum wages (Card, 1982, p. 52). The complications with quasi-experimental evaluations have led some researchers to question the results in these papers (Neumark and Wascher, 2000, p. 1362). Even though some researchers find the results questionable, the classical theory on minimum wage was contradicted by their results.
3. Descriptive Statistics
3.1. Minimum wages and youth employment and youth unemployment in the European Union
Teenagers and youth have consistently participated less than adults in the labour supply. One explanation is that a majority of youth assigns their time in school and leisure activities. Nevertheless, there is a proportion that does participate in the labour supply. In 2014, the average youth unemployment in the European Union was 22.2%, and more than double than the unemployment rate among adults. Thus, there has been growing concerns about the functioning of the youth labour market in the European union, youth unemployment remained high and thus became a major issue for the member states of the European Union.
9
Chart 2: average unemployment rate of the European Union Memver States by age group in percentages. Source: Eurostat
Chart 2 displays the average unemployment rate by age group in the years 2000 to 2014 of the European Union member states. It illustrates the annual average unemployment rate in percentages of the total labour supply per age group. The chart shows that the youth unemployment rate in comparison to the adult and the total unemployment rate is high. This reveals a serious problem within the European Union and asks for a policy relevant solution.
Chart 3: youth unemployment rate per country in the year 2014. Source: Eurostat
Chart 3 displays the youth unemployment rate of the different states in the European Union in 2014. It indicates that there are big differences in youth unemployment rate between the
5 7 9 11 13 15 17 19 21 23 25 2000 2002 2004 2006 2008 2010 2012 2014
Youth unemployment Adult unemployment Total unemployment
0.0 10.0 20.0 30.0 40.0 50.0 60.0 G erm an y Au stria Malta De n m ar k N eth erl an d s Es to n ia Cze ch Rep u b lic U n ite d K in gd o m Lith u an ia La tv ia Slov en ia H u n gary Finla n d Av era ge E u ro p e an U n ion Lu xe m b o u rg Sw ed e n Be lgi u m Bu lgari a Ire lan d Po lan d Rom an ia Fran ce Slov ak ia Po rtu gal Cyp ru s It aly C roa tia G re e ce Sp ain
10
countries, ranging from 7.7 to 53.2 percentages. The average unemployment rate in the European Union is 22.2 percent.
In this paper the influence of the different minimum wage regulations on the relative position of youth employment to total employment and the relative position of youth unemployment to total unemployment is examined. Twenty-one of the twenty-eight European Union Member States have some kind of statutory minimum wage. There are commonalities and differences in the extent to which minimum wages have established a common floor for wage levels in the different member states (Human rights report, 2014). Some of the member states implement lower minimum wage rates for younger and less experienced workers, this is an age-dependent minimum wage. The countries which implement a national minimum wage without age-based different rates are Bulgaria, Croatia, Estonia, France, Greece, Hungary, Latvia, Lithuania, Poland, Portugal, Romania, Slovenia, and Spain. The countries which implement a national minimum wage with age-based different rates are Belgium, Czech Republic, Ireland, Luxembourg, Malta, the Netherlands, Slovakia, and the United Kingdom. Each of these country applies a different strategy to
differentiate for younger and less experienced workers.
Countries Age 15 16 17 18 19 20 21 22 23 Belgium Czech Republic Ireland Luxembourg Malta Netherlands Portugal Slovakia United Kingdom 70 80 70 75 94.2 30 75 50 58.3 70 80 70 75 94.2 34.5 75 75 58.3 76 80 70 80 95.9 39.5 75 75 58.3 82 80 83 100 100 45.5 100 75 78.9 88 90 83 100 100 52.5 100 75 78.9 94 90 83 100 100 61.5 100 100 78.9 100 90 83 100 100 72.5 100 100 100 100 100 83 100 100 85 100 100 100 100 100 100 100 100 100 100 100 100 Table 1: Youth minimum wage per age group in percentages of national minimum wage per country. Source: EIRO
Table 1 gives a brief overview of the youth minimum wage per age group in percentages of national minimum wage per country. The table illustrates that there are big differences in the level of differentiation for the age-dependent minimum wages in the European Union. Standing out is the Netherlands, this country implements the most differentiated minimum
11
wage system in the European Union. It is unique based on a change in percentage each year of age. The relative youth unemployment rate in the Netherlands belongs to one of the lowest rates of the European Union. This may suggest that an age-dependent minimum wage softens the negative employment effects of a national minimum wage.
The linear relation between the minimum wage and the relative youth unemployment and the relative youth unemployment is illustrated in chart 4 and chart 5. It shows a scatterplot on the group of countries that implement a national minimum wage in the years 2010 to 2014. The relative position of youth to the total population is used, it allows to ignore macroeconomic developments affecting aggregate employment and unemployment (van Soest, 1994, p. 101).
Chart 4 suggest that there is a quite steep positive relationship between the relative youth employment and the minimum wage. This implies that when minimum wage is high the relative youth employment will also be high. This is in contradiction to the classical theory on minimum wages, it specifies that if a country implements a high minimum wage that relative youth employment will be low. The line of best fit between the relative youth unemployment and the minimum wage is also quite steep and positive. This implies that when the minimum wage is high, the relative youth unemployment will also be high. This in in agreement with the classical theory on minimum wages. Remarkable is that chart 4 and 5 show contradictory relations with regard to the classical theory. According to chart 4 and 5, higher minimum wages indicates higher relative youth employment and higher relative youth unemployment. Thus, an increase in the youth labour supply after an increase in the minimum wage can be expected according to the linear relationship in charts 4 and 5.
Further, chart 4 and 5 suggest that there are differences between rich and poor countries with regard to the relationship between minimum wage and youth employment and youth unemployment. Countries which implement a minimum wage below 1000 euros a month shows a different relationship with regard to the youth employment and youth unemployment. The best fit line will slope positive for countries implementing a minimum wage below 1000 euros a month, and the best fit line may slope negative if only the countries which are considered are the countries with a minimum wage above 1000 euros a month.
12
Chart 4 and 5: linear relationship between the minimum wages a month in euros and the relative youth employment or relative youth unemployment in percentages for the European Union. Source: Eurostat
4. Methodology and Data
4.1 Methodology
The influence of the different minimum wage regulations on youth employment, youth unemployment, and the youth labour supply is tested using a cross-sectional analysis between the 28 European Union countries. The 28 countries are classified in three different groups. First, the 7 countries without a statutory minimum wage. Second, the 12 countries that implement a statutory minimum wage without age-dependent regulations. Third, the 9 countries that implement a statutory minimum wage with age-dependent regulations. The cross-sectional analysis will be examined on three different time periods.
In this study the focus is on the relative position of youth compared with the full population in that country. Thus, in this cross sectional analysis, the youth employment is taken as a percentage of total employment, the youth unemployment is taken as a percentage of total unemployment, and the youth labour supply is taken as a percentage of the total labour supply. The labour supply is total employment plus total unemployment. Comparing the relative position of youth to the full population allows to ignore macroeconomic developments affecting aggregate employment and unemployment. It allows to ignore the business cycle. The business cycle is the short-run fluctuations in output, incomes, and employment in the economy (Mankiw, 2013). These are substantial fluctuations around the
13
average GDP growth in an economy that can occur in the short-run.
In this cross-sectional analysis the independent variables minimum wage and age-dependent minimum wage are represented as dummy variables in the model. Dummy variable 1 is MINWAGE, it takes the value 1 if a country implements a statutory minimum wage and the value 0 if a country does not implement a statutory minimum wage. Dummy variable 2 is AGEDEP, it takes the value 1 if a country implements a statutory minimum wage with age-dependent regulations and it takes the value 0 if a country implements a statutory minimum wage without age-dependent regulations. In formal terms, the model specifications for this study is:
𝑌𝑖 = 𝛽0+ 𝛽1𝑀𝐼𝑁𝑊𝐴𝐺𝐸 + 𝛽2𝐴𝐺𝐸𝐷𝐸𝑃 + 𝜀𝑖 (1) 𝑌𝑖 = 𝛽0+ 𝛽1𝑀𝐼𝑁𝑊𝐴𝐺𝐸 + 𝛽2(𝐴𝐺𝐸𝐷𝐸𝑃 𝑋𝑖) + 𝜀𝑖 (2) 𝑌𝑖 = 𝛽0+ 𝛽1𝑀𝐼𝑁𝑊𝐴𝐺𝐸 + 𝛽2(𝐴𝐺𝐸𝐷𝐸𝑃 𝑋𝑖) + 𝛽3𝐸𝐷𝑈𝑖+ 𝛽4𝐺𝐷𝑃𝑖+ 𝜀𝑖 (3) With 𝑌𝑖 = { 𝑅𝑌𝐸 𝑅𝑌𝑈 𝑅𝑌𝐿𝑆 𝑅𝑌𝐸/𝑅𝑌𝐿𝑆 𝑅𝑌𝑈/𝑅𝑌𝐿𝑆
The variables of interest are RYE, RYU, RYLS, RYE/RYLS, and RYU/RYLS as the dependent variables. RYE is the relative youth employment, RYU is the relative youth unemployment, RYLS is the relative youth labour supply, RYE/RYLS is the relative youth employment divided by the relative youth labour supply, and RYU/RYLS is the relative youth unemployment divided by the relative youth labour supply. The final two dependent variables are added because of the indirect effect of the youth labour supply on youth employment and youth unemployment. The indirect effect shows that the proportion of youth in the population is negatively related to youth employment. If the proportion of youth in the population is high, youth employment will be low and vice versa. The same indirect effect is true for the youth labour supply on youth unemployment. If the proportion of youth in the population is high, youth unemployment will be high and vice versa.
Equation (1) examines the relationship between each dependent variable and the different minimum wage regulations of the member states in the European Union. In equation (2) the variable 𝑋𝑖 is added. 𝑋𝑖 is the main explanatory variable of interest, it states the degree of differentiation of an age-dependent minimum wage in a country. 𝑋𝑖 defines the different levels of age-dependent minimum wages and takes a value between 0 and 1. If a
14
country implements a highly differentiated age-dependent minimum wage the variable 𝑋𝑖 will be close to 1, and if a country has a small differentiated age-dependent minimum wage system the variable 𝑋𝑖 will be close to 0. Thus, 𝑋𝑖 standardizes the level of differentiation of an age-dependent minimum wage. This allows for comparisons between the countries who implement an age-dependent minimum wage. 𝑋𝑖 is calculated by taking the average of the minimum wage percentages for the age 15 to 23 of each country in table 1. The average age-dependent minimum wage percentages of each country is than standardized between the value of 0 and 1 with use of equation (4). MINWAGEPERC is the average age-dependent minimum wage percentage of country (𝑖).
𝑋𝑖 =100−𝑀𝐼𝑁𝑊𝐴𝐺𝐸𝑃𝐸𝑅𝐶(𝑖)50 (4)
In equation (3) two more control variables are added. EDU is the educational attainment level for youth in each country and GDP is the gross domestic product growth rate in market prices of the last 4 years in each country. Education is added as a control variable because it is widely believed that labour market failures amongst youth is influenced by the educational attainment level (Ingham, 1989, p. 5). The GDP growth rate is added as a control variable because theory suggests that youth employment and youth unemployment is highly sensitive to the strength of the economy as a whole. When the level of economic activity is high in a country, youth employment will also be high (Freeman & Wise, 1982, p. 3).
According to the classical theory on minimum wages, implementing a minimum wage has negative employment effects and implementing an age-dependent minimum wage softens the negative employment effects (Mankiw, 2013, pp. 181-184). However, previous research has not been able to come to a consensus. Previous research finds ambiguous results on the youth employment effects with regard to the different minimum wage implementations. Based on the contradiction in previous empirical literature the hypothesis is formed. The expectation is that there is some sort of significant correlation between the minimum wage and the youth employment and youth unemployment. Thus, the parameters are expected to be shaped according to the restrictions in table 2.
15
Yi Equation (1) Equation (2) Equation (3)
RYE 𝛽1≠ 0, 𝛽2≠ 0 𝛽1≠ 0, 𝛽2≠ 0 𝛽1≠ 0, 𝛽2≠ 0, 𝛽3≠ 0, 𝛽4≠ 0 RYU 𝛽1≠ 0, 𝛽2≠ 0 𝛽1≠ 0, 𝛽2≠ 0 𝛽1≠ 0, 𝛽2≠ 0, 𝛽3≠ 0, 𝛽4≠ 0
RYLS 𝛽1≠ 0, 𝛽2≠ 0 𝛽1≠ 0, 𝛽2≠ 0 𝛽1≠ 0, 𝛽2≠ 0, 𝛽3≠ 0, 𝛽4≠ 0 RYE/RYLS 𝛽1≠ 0, 𝛽2≠ 0 𝛽1≠ 0, 𝛽2≠ 0 𝛽1≠ 0, 𝛽2≠ 0, 𝛽3≠ 0, 𝛽4≠ 0 RYU/RYLS 𝛽1≠ 0, 𝛽2≠ 0 𝛽1≠ 0, 𝛽2≠ 0 𝛽1≠ 0, 𝛽2≠ 0, 𝛽3≠ 0, 𝛽4≠ 0 Table 2: Hypothesis per equation and per dependent variable in formal terms
4.2 Data
This study looks at the influence of the different minimum wage regulations on relative youth employment, relative youth unemployment, and the relative youth labour supply of the 28 countries in the European Union. In this cross-sectional analysis three different time periods are examined: before the recession (2006), during the recession (2010), and after the recession (2014). The data on the 28 countries are gathered from Eurostat, the statistical office of the European Union. Youth consists of the persons in the age group 15 to 24 and the total population consists of the persons in the age group 15 to 74.
The data on youth employment and total employment is gathered from the employment by sex, age, and citizenship theme. The data on youth unemployment and total unemployment is gathered from the unemployment by sex, age and citizenship theme. The data on youth labour supply and total labour supply is gathered from the active population by sex, age and citizenship theme. The data on education is gathered from the young people by educational attainment level theme. The tertiary education is used for education, refers to any type of education pursued after the high school level. The GDP growth rates are calculated using the GDP at market prices theme. The GDP growth rate over the last 4 year is taken. For the data on age-dependent minimum wages see table 1.
5. Results
Table 3 shows the results of regression (1) and (2). Each dependent variable will be discussed and compared for the two regressions.
In regression (1), it appears to be that with RYE as dependent variable the minimum wage shows significant negative effects. This correlation is in agreement with the classical theory, if a country implements a minimum wage the youth employment will decrease relative to total employment. The implementation of an age-dependent minimum wage is
16
positive and very small insignificant. The positive correlation is again in agreement with the theory, an age-dependent regulation softens the negative youth employment effects of an minimum wage. In regression (2), the minimum wage dummy gives again negative effects but also shows one very small insignificant effect in 2006. The insignificant result in 2006 may be the result of a period of high economic growth. When 𝑋𝑖 is added in the model, the age-dependent coefficient appears to be positive and very significant. This correlation suggests that a more differentiated age-dependent minimum wage will soften the negative youth employment effect of a minimum wage. It also implies that the level of differentiation of an age-dependent minimum is an important factor considering the employment effects.
Regression (1) and (2) with RYU as dependent variable shows some mixed results. In both regressions the dummy variable MINWAGE shows negative significant results, with exception for the year 2006 a negative insignificant effect. The negative correlation implies that countries with a national minimum wage will have smaller relative youth unemployment rates than countries without a national minimum wage. This is in contradiction to the classical theory on unemployment. In regression (1) the age-dependent dummy shows positive insignificant effects on RYU with as exception in 2014 a small significant effect. While it is mostly insignificant, the positive correlation could still imply that if a country uses an age-dependent minimum wage that relative youth unemployment will increase. This is again in contradiction to the classical theory. In regression (2) with the added coefficient 𝑋𝑖 as a measurement of the level of differentiation with age, the years 2010 and 2014 show significant results. The insignificant result in 2006 may be the result of an period of high economic growth. The results on correlation between an age-dependent minimum wage on RYU remains positive when 𝑋𝑖 is added, still in contradiction to the classical theory.
The effect on the RYLS of implementing a national minimum wage in a country shows in both regressions a negative correlation. The correlation shows that a country with a national minimum wage will have a smaller youth labour supply relative to total labour supply. This implies that youth will participate less in the job market if a country has a minimum wage, youth will hold less jobs or are less eager in searching for a job. This coincides with the classical theory, the lowest quality of labour and less experienced people will be excluded from the labour market with a minimum wage. While the negative effects coincide with the theory, for the year 2006 the effects are insignificant for both regressions. This insignificant effect may be the result of an period of high economic growth. In regression (1) the dummy variable AGEDEP shows positive but insignificant effects. While
17
the result is insignificant, the positive correlation still coincides with the theory. It may suggests that a country which differentiates the minimum wage for age-groups will have an increase in the relative youth labour supply. The age-dependent minimum wage may encourages youth to participate in the job market. In regression (2) the coefficient remains positive but with the addition of 𝑋𝑖 in the model the results are very significant. The level of differentiation is an important factor considering the relative youth labour supply.
Regression (1) and (2) with RYE/RYLS as dependent variable accounts for the indirect effect of the youth labour supply on the relative youth employment. Regression shows that implementing a national minimum wage has a negative insignificant correlation with regard to the RYE/RYLS. With exception for the year 2010 where the effect is small significant, nevertheless still negative. This exception may be explained by the global economic crisis in 2010. While the effects are mostly insignificant, it may indicate that countries with a minimum wage have negative relative employments effect. This coincides with the classical theory. The coefficient AGEDEP shows insignificant positive results in regression (1), with again as exception a significant result for the year 2010. The positive effects may indicate smaller negative youth employment effects for countries with age-dependent minimum wages. Addition 𝑋𝑖 in regression (2) shows a positive significant correlation. This implies that the level of differentiation in age is an important factor with regard to the relative youth employment. The positive correlation is in agreement with the classical theory, an age-dependent minimum wage softens the negative employment effects. The regressions on RYU/RYLS shows very mixed results. With this regression the indirect effect of the labour supply on youth unemployment is accounted for. Regression (1) shows insignificant results for MINWAGE and AGEDEP. The coefficient of MINWAGE is positive for 2006 and 2014 and negative for 2010. The positive effect is in consensus with the theory, if a country implements a minimum wage there will be more relative youth unemployment. The negative effect in 2010 is in contradiction with this theory, this may be explained by the global economic crisis. The coefficient of AGEDEP is positive for each year, even though it is insignificant it may imply that countries with an age-dependent minimum wage have increased youth unemployment. This is in contradiction to the theory. In regression (2), the year 2010 shows significant results the other results are insignificant. The exception for year 2010 may be related to the economic recession in the European Union. It may suggests that having some form of age-dependent minimum wage could be good to avoid high increases in youth unemployment during economic crisis. While the results for
18
2006 and 2014 are insignificant, the coefficient of (AGEDEP 𝑋𝑖) shows large negative correlation. The negative effect hints that countries with a highly age-dependent minimum wage will have lower relative youth unemployment, which is in agreement with the theory.
Table 4 shows the results of regression (3). The results clearly demonstrates that the control variable educational attainment level can be omitted from the model. The coefficient EDU has only insignificant results. While the coefficient is insignificant, EDU shows a small positive effect for the dependent variables RYE, RYU, and RYLS. This may indicate that countries with a higher educational attainment level will possibly have higher relative youth employment, but also higher relative youth unemployment, and a higher level of youth in the labour supply. The results for RYE/RYLS and RYU/RYLS show insignificant mixed results. The added control variable GDP growth rate shows mixed results. Most of the results are insignificant, with exception to the year 2014 where there are significant results for the dependent variables RYLS and RYE/RYLS. The coefficient GDP is close to zero with respect to the dependent variables. The most remarkable result of regression (3) is: with the addition of two control variables the results of MINWAGE and AGEDEP are less significant. It also shows more mixed results according to the coefficients.
6. Conclusion and discussion
6.1 Conclusion
This study aimed to investigate the youth employment and unemployment effects of the differences in minimum wage regulations in the European Union. The different minimum wage regulations within the European Union can be classified in three groups: countries without a national minimum wage, countries with a national minimum wage, and countries with a differentiated national minimum wage that is age-dependent. To examine this effect a cross-sectional analysis between the 28 European Union countries in three different time periods were conducted. The relative youth employment, relative youth unemployment, and relative youth labour supply was chosen as dependent variables. Results of the regressions were not always conclusive. However, no strong results were expected in this research as the population is small (28 countries). The regressions concluded two interesting results.
19
Regression (1) Regression (2)
Constant MINWAGE AGEDEP R-squared Constant MINWAGE (AGEDEP Xi) R-squared
RYE 2006 2010 10.94118 (0.00)*** 10.49621 -2.076227 (0.047)* -3.125217 2.588156 (0.082) 2.255658 0.1203 0.2179 10.94118 (0.00)*** 10.49621 -1.972049 (0.09) -3.119799 6.789027 (0.00)*** 6.493569 0.2492 0.3138 (0.00)*** (0.01)** (0.098) (0.00)*** (0.018)* (0.001)*** 2014 9.939379 -3.576223 2.303476 0.2377 9.939379 -3.600799 6.834613 0.3381 (0.00)*** (0.01)** (0.094) (0.00)*** (0.014)* (0.002)** RYU 2006 26.65794 -3.348752 4.975366 0.1203 26.65794 -2.211647 6.722582 0.0532 (0.00)*** (0.233) (0.107) (0.00)*** (0.459) (0.221) 2010 26.9083 -6.012274 3.9429 0.1789 26.9083 -5.920262 10.79322 0.233 (0.00)*** (0.038)* (0.135) (0.00)*** (0.042)* (0.007)** 2014 25.43835 -7.456369 5.4894101 0.2477 25.43835 -6.694066 10.74165 0.2287 (0.00)*** (0.016)* (0.048)* (0.00)*** (0.029)* (0.042)* RYLS RYE/RYLS RYU/RYLS 2006 2010 2014 2006 2010 2014 2006 2010 2014 11.9354 (0.00)** 11.73126 (0.00)*** 11.38449 (0.00)*** 91.16729 (0.00)*** 88.69631 (0.00)*** 85.49646 (0.00)*** 229.1063 (0.00)*** 237.3916 (0.00)*** 230.3813 (0.00)*** -1.83842 (0.063) -2.70733 (0.021)* -3.59772 (0.003)** -3.344923 (0.195) -7.060703 (0.025)* -4.195272 (0.432) 2.932965 (0.913) -4.441632 (0.871) 0.7764015 (0.978) 2.474888 (0.1) 1.894278 (0.17) 2.104041 (0.111) 2.519116 (0.217) 5.746847 (0.021)* 4.766239 (0.215) 1.334473 (0.948) 3.392637 (0.856) 17.97516 (0.471) 0.1525 0.1628 0.2606 0.0919 0.2596 0.0568 0.0012 0.0018 0.0251 11.9354 (0.00)*** 11.73126 (0.00)*** 11.38449 (0.00)*** 91.16729 (0.00)*** 88.69631 (0.00)*** 85.49646 (0.00)*** 229.1063 (0.00)*** 237.3916 (0.00)*** 230.3813 (0.00)*** -1.685544 (0.134) -2.76908 (0.03)* -3.648315 (0.005)** -3.544484 (0.147) -6.290431 (0.041)* -3.903749 (0.441) 10.32665 (0.701) -0.4057377 (0.988)** 11.6085 (0.675) 6.132194 (0.001)*** 5.901111 (0.001)*** 6.432998 (0.002)** 8.640842 (0.007)** 11.434 (0.003)** 11.82909 (0.033)* -46.08132 (0.137) -17.44088 (0.551)** -21.13286 (0.652) 0.204 0.2654 0.3666 0.1738 0.2415 0.0745 0.0405 0.0068 0.0110 The coefficients between the parenthesis are the matching p-values. Coefficients followed by one, two, and three stars are significantly different from zero at the 10%,
5%, and 1% level, respectively.
20 Regression
(3)
Constant MINWAGE (AGEDEP Xi) EDU GDP R-squared
RYE 2006 2010 10.4174 (0.00)*** 10.0.806 -1.864958 (0.257) -3.213067 5.753365 (0.007)** 5.75909 0.1502995 (0.148) 0.107342 -.0099367 (0.709) -0.0099392 0.3092 0.3419 (0.00)*** (0.018) (0.014) (0.473) (0.792) 2014 8.485408 -4.022589 6.622074 0.0943712 0.0938986 0.5136 (0.00)*** (0.003)** (0.011)* (0.367) (0.002) RYU 2006 26.05016 3.806613 3.629191 0.3238189 -0.494097 0.1376 (0.00)*** (0.744) (0.467) (0.212) (0.535) 2010 24.49055 -6.646706 8.237522 0.460971 0.0049795 0.3286 (0.00)*** (0.037)* (0.012)* (0.122) (0.949) 2014 21.96957 -7.328246 9.656333 0.2991678 0.1395879 0.3492 (0.00)*** (0.022)* (0.017)* (0.297) (0.1) RYLS RYE/RYLS RYU/RYLS 2006 2010 2014 2006 2010 2014 2006 2010 2014 11.52964 (0.00)** 11.16696 (0.00)*** 9.853013 (0.00)*** 89.55693 (0.00)*** 88.72219 (0.00)*** 83.21519 (0.00)*** 235.7512 (0.00)*** 230.0334 (0.00)*** 239.1182 (0.00)*** -1.468029 (0.380) -2.860746 (0.039)* -3.935785 (0.002)** -4.484049 (0.089) -6.520374 (0.029)* -6.092844 (0.145) 16.02398 (0.543) -4.44646 (0.864) 11.50446 (0.681) 1.972328 (0.018)* 4.865665 (0.024*) 5.965448 (0.015)* 8.348814 (0.019)* 12.80113 (0.008)** 13.86914 (0.008)* -49.02553 (0.116) -14.90598 (0.620) -15.75522 (0.761) 0.1406591 (0.189) 0.1426617 (0.358) 0.1305959 (0.202) 0.2336571 (0.121) -.1119814 (0.668) -.1557707 (0.632) -0.6364214 (0.675) 0.5789248 (0.749) -1.091985 (0.616) -0.0138393 (0.624) -0.0179318 (0.614) 0.0633251 (0.028)* .0273638 (0.545) .0582255 (0.626) 0.4938722 (0.00)*** -0.1960006 (0.679) 0.4637585 (0.402) 0.0344563 (0.968) 0.2653 0.3227 0.4884 0.2193 0.2649 0.5074 0.0485 0.0211 0.0179 The coefficients between the parenthesis are the matching p-values. Coefficients followed by one, two, and three stars are significantly different from zero at the 10%, 5%, and 1% level, respectively.
Table 4: results regression (3)
First, the regressions showed mostly significant negative effects on the relative youth employment, relative youth unemployment, and relative youth labour supply by implementing a national minimum wage. This correlation implies that if a country implements a national minimum wage that the youth employment, youth unemployment, and the youth labour supply will decrease. The decrease in employment and labour supply is in agreement with the classical theory on minimum wages, the decrease in unemployment is in contradiction to the classical theory.
Second, the regressions showed that the level of differentiation for an age-dependent minimum wage is mostly significant and has a positive youth employment effect, positive youth unemployment effect, and a positive youth labour supply effect. The correlation could imply that a more differentiated minimum wage will result in an increase in youth employment, youth unemployment, and youth labour supply. Again, the youth employment
21
and youth labour supply effect is in agreement with the classical theory and the youth unemployment effect is in contradiction to this theory.
6.2 Discussion
An important finding of this research is the relationship between the variable 𝑋𝑖 and the dependent variables. The research shows a relationship between the degree in differentiation of an age-dependent minimum wage on youth employment and youth unemployment. Further research could elaborate this relationship even more by emphasising the following suggestions.
First, the fact that the results of this study are not always conclusive is caused by the small number of observations. No strong results were expected in this research as the population is small. Relative small number of observations results more quickly in insignificant results. In this study only 28 countries are in the sample, with three different time periods conducted. Research into the effects of minimum wage is often examined with long-term time series data. Future studies looking at the effect of minimum wage regulations on youth employment might get better results by using a bigger population.
Second, the minimum wage effects may be underestimated by only considering youth employment, youth unemployment, and the youth labour supply. People who are not actively searching for a job are not included in the official unemployment count (Brown, Gilroy and Kohen, 1982, pp. 19-20). Thus, the unemployed who are not actively searching for a job are not included in this study. Future studies may want to include this population group for better results.
Third, with this study three different time periods were examined. Looking at the results, the study find some different relationships in the different periods. This may be explained by the differences in economic growth in the three time periods. 2006 was characterized by a period of high economic growth, 2010 was characterized by a period of economic crisis and low economic growth, and 2014 is characterized by small improvements in economic growth. Further research could emphasise more on the relationship of an age-dependent minimum wage and the relative position of youth in different time periods with regarding to economic growth.
22
7. Literature
Brown, C., Gilroy, C. & Kohen, A. (1982) The effect of the minimum wage on employment and unemployment, Journal of Economic Literature, 20(2), 487-528.
Card, D. (1982) Do minimum wages reduce employment? A case study of California, 1987-1989, Industrial Relations & Labor, 46(1), 38-54.
Country reports on human rights practices (2014) Human Rights, retrieved from: http://www.humanrights.gov/dyn/countries.html
Freeman, R. & Wise, A. (1982) The Youth Labor Market Problem: Its Nature, Causes, and Consequences, University of Chicago Press, 1-16.
Gomez-Salvador, R. & Leiner-Killinger, N. (2008) An analysis of youth unemployment in the euro area, European Central Bank: Occasional Paper Series, 89, 2-45.
Hyslop, D. & Stillman, S. (2005) Youth minimum wage reform and the labour market in New Zealand, Labour Economics, 14(2), 201-230.
Ingham, M. (1989) Education and Youth Unemployment: A Reappraisal, Journal of
Economic Studies, 16(3)
Mankiw, G. (2013) Macroeconomics, (edition), city: Publiher.
Neumark, D. & Wascher, W. (2000) Minimum Wages and Employment: A Case Study of the Fast-Food Industry in New Jersey and Pennsylvania: Comment, The American Economic
Review, 90(5), 1362-1396.
Neumark, D. & Wascher, W. (2004) Minimum wages, labor market institutions, and youth employment: a cross national analysis, Industrial Relations & Labor, 57(2), 223-248.
Rebitzner, J. & Taylor, L. (1995) The consequences of minimum wage laws: Some new theoretical ideas, Journal of Public Economics, 56(2), 245-255.
Stock, J. & Watson, W. (2012) Introduction to Econometrics, (edition), city: Publisher
United Nations for statistical purposes (2008), retrieved from:
23
Van Soest, A. (1989) Minimum wage rates and unemployment in the Netherlands, De
Economist, 137(3), 279-308.
Van Soest, A. (1994) Youth minimum wage rates: the Dutch experience, International
Journal of Manpower, 15(2/3), 100-117.
Vuuren, D. & Bosch, N. (2012) Het wettelijk minimumjeugdloon en de arbeidsmarkt voor jongeren, Centraal Planbureau, 1-16.
