The effect of minimum wage on unemployment in Western Europe

(1)

The effect of minimum wage on unemployment in

Western Europe

Amsterdam Business School

Name

Laura Schim van der Loeff

Student number 10016155

Program

Economics & Business

Specialization

Economics

Number of ECTS 12

Supervisor

N. Ciurila

Date

April 2016

(2)

1

1. Statement of Originality

This document is written by Laura Schim van der Loeff, 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.

(4)

3

2. Introduction

With the introduction of a minimum wage in Germany in January 2015, controversy arose around this new law. Many feared that Germany’s economic position would be weakened, by rising unemployment and companies moving their business to lower-waged countries, among other things. The debate on minimum wage is not one that has not happened before, on numerous occasions. When the United Kingdom implemented a minimum wage in 1999, the same concerns arose. But not only the implementation of a minimum wage sparks up the debate again. Also rising the minimum wage leads to the same heated discussion, since there seems to be little consensus about the different effects the minimum wage might have. Though a lot of research has been done about it in the United States, there has been less research on the topic in Europe. Having a less ambiguous answer about the effects of a minimum wage is crucial for policy makers, in order to take adequate decisions about it.

Since the 1st_{of January 2015, there are only 6 European Countries left who do not have a minimum} wage (Hanke, 2014). Will it be a matter of time before those other countries follow as well? There are numerous reasons why people are in favor of a minimum wage, as there are multiple arguments from those who oppose such a wage. The supporters of a minimum wage often point to the

reduction of poverty and inequality, and the increased standard of living of workers. On the other hand, the opponents mention the increase in poverty and unemployment, and the damage it does to businesses. In this bachelor thesis I will be focusing on the unemployment argument, where my research question is: What have been the effects of a minimum wage on unemployment in Europe from 2000-2014?

To answer this question I will do an econometric analysis based on one done by Neumark and Wascher (2004). I will do a pooled analysis of 8 western-European countries, trying to keep as many similarities between them. Both the effects of minimum wage on both unemployment and on youth unemployment will be examined, since some theories and previous evidence predict a minimum wage might affect them differently. Though no significant effect of a minimum wage on

unemployment was found, there was enough statistical evidence to suggest that a minimum wage raises youth unemployment.

This introduction is succeeded by a literature review. Besides theory, some of the previous literature and empirical findings will be discussed with respect to this topic. Consequently, I will explain my

(5)

4 methodology, followed by my hypotheses. Thereafter , the empirical results will be given along with their interpretation. Finally, a conclusion will be presented followed by a discussion and limitations.

3. Literature Review

Though a lot has been said, researched and written about minimum wage, a satisfactory answer has yet to be found. Not only theories, but also empirical research seem unable to find a consensus on the effect it has on unemployment, despite the plethora of research. In this literature review I will first discuss two theories on minimum wage and its effects on unemployment, followed by evidence from previous research.

3.1 Theory

The first theory, the classical view, states that a minimum wage will have an adverse effect on unemployment, and with that hurting the economy. This is under the assumption of perfect competition, where the ‘equilibrium wage’ will be set at the intersection of the demand and supply curve. When the minimum wage is set higher than the equilibrium market wage, a difference in supply and demand will be created, resulting in unemployment (Pindyck & Rubinfield, 2009). The rationale behind the classical theory is that an increase in minimum wage will lead to an increase in production costs, resulting in a higher price. This higher price will lead to less demand, therefore less production, and less demand for labor. Another reason for declining demand, as described by Machin & Manning (1996, p.668), is the effect of low-wage workers being shut out, for example if companies chose to take their labor elsewhere, to countries with lower wages, or invest in technology which would make low-paid workers redundant.

As shown in Figure 1, the equilibrium wage W1* results in an amount of Labor Q1*. By setting the minimum wage at a higher level W2, Labor Demand will fall back to Q2, there will be less hiring and more firing while supply will rise to Q3, more people will be willing to work at the new wage level, leading to an unemployment level of Q3-Q2 .

(6)

5 Figure 1: Neoclassical Minimum Wage Model

However, the model explained above works under the assumption of a perfect competition. If that assumption would not hold and the economy would be a monopsony, a minimum wage could actually increase employment. A monopsonic market assumes that there is only one buyer of labor (the employer) to multiple sellers (employees)(Dolado et al, 1996, p. 329). According to theory, if the labor market is a monopsony, the employer will set a demand for labor at the intersection of their Marginal Cost (MC) and Marginal Revenue (MR) curve, and ’natural’ wage (the theoretical wage that would arise from supply and demand) and employment will both be lower than in perfect

competition. The employer has more power, since there are few employers, and a lot of people are willing to work, so it can set a relatively low wage. By increasing the minimum wage, the employer would be forced by the government to pay their workers more, but it will still be profitable to pay them this amount. They will not need to employ less people, so there will be no increase in

unemployment. In Figure 2, point Q1 , corresponding with wage W1, is the intersection between the Demand and the Supply curve, would be the employment and the wage in case of a competitive market. In a monopsony, the quantity would be set at Q2, where the profit is maximized at MC = MR by the employer. The wage corresponding with that would be W2. Raising the minimum wage up to W3, would therefore not cause an increase in unemployment, since the employer will still chose the same amount of labor, since it is still making profit. .

(7)

6 Figure 1: Monopsonic Minimum Wage Model

According to Dolado et al. (1996, p. 330), most textbooks argue that the monopsonic model is highly unrealistic and extremely rare. However, they contest that, stating that the most important features of this model hold in a situation where a firm has some discretion over the wages they pay. They argue that perfect competition is the unrealistic outcome, not monopsony. As Neumark and

Wascher (2006, p. 123) point out, it would be useful for the literature to bring more direct evidence as to which model characterizes the low-wage market in a more correct way.

3.2 Empirical Findings

The evidence on the effects of a minimum wage on unemployment are, like the theory, ambiguous. Though more than a 100 studies have been published about the possible effects of a minimum wage, there is still no clear consensus (Neumark and Wascher, 2006, p. 5). Though most research done before the 1990’s suggests a negative effect between employment and minimum wage, the general view is that recent research done in later years (known as the “new minimum wage research” seems to fail to support this view more often than not (Neumark and Wascher, 2006, p. 4).

The research done so far can be roughly assigned to one of three categories: time-series analysis, cross-sectional (panel) analysis, or a ‘natural experiment’, researching effects after a sudden shock to minimum wage (sudden increase/decrease). This literature review will focus on the last two, since, as Neumark and Wascher (2004, p. 223) point out, there are serious limitations to time-series analysis. Also, most of the older research has been done with time-series variation, however, Neumark and Wascher (2004, p. 223) note that there is a shift towards cross sectional or panel data

(8)

7 to identify the effects, and this literature review will focus on the more recent research (post 1990). First, evidence from the United States will be discussed, followed by findings in Europe.

3.2.a. Empirical Findings in the United States

Most of the empirical evidence has its origin in the United States. First, evidence supporting a negative impact of minimum wage on employment will be discussed, followed by the evidence contradicting this view.

Neumark and Wascher (1992) researched the effects of unemployment on teenagers in the United States, young adults (up to 24), and controlled for year effects. As a control variable they used men above the age of 25. They found evidence supporting the earlier view of minimum wage effects. The authors researched the effects once more in 1996 and 1995, including school enrolment as a

variable, concluding that, consistent with their hypothesis, the least-skilled teenagers are the ones who feel the negative impact of minimum wage the most. In 2003, they repeated their research, and found similar results (Neumark and Wascher, 2006, p. 24). Some other research which supports the theory of negative minimum wage effect was conducted by Lang & Kang (1998); who found

negative employment effects of the minimum wage on adults, they discovered that the employment is shifted from adults to teenagers, raising employment with the latter, but leading to a decrease of employment among adults. The rise in employment among teenagers is greater than the decrease in employment of adults, therefore as a whole, there might be positive effects. The authors emphasize that the consequences of a minimum wage raises concerns, even though the employment effects are beneficial.

Not only direct effect on employment, but also the effect of minimum-wage on hours worked has been researched. Zavodny (2000) researched the effects on teen-unemployment in the United States. She also observed the effects on the amount of hours worked, contesting that it might be better for companies in the short run to lower the amount of hours, instead of their amount of employees. According to the author, this would explain why some research might find no negative effects on unemployment due to a rise in minimum wage, when there was in a fact a negative effect on hours worked. She found that an increase in minimum wage led to slight negative effects on teen-employment, but the hours of work were not reduced.

One of the first, and most famous evidence found against the general view, was offered by Card and Krueger (1994), researching the effects of a minimum wage on the employment in the fast-food industry in New Jersey, though this research has been heavily debated. They compared the unemployment in the fast-food industry in New Jersey, after a rise in minimum wage, to

(9)

8 Pennsylvania, where the wage had not increased. They did not find a negative effect, on the

contrary, they measured a larger increase in employment in New Jersey than in Pennsylvania. They also compared stores unaffected by the minimum wage (the wages in those stores were already higher than the new minimum), but found the same increase in both stores, suggesting that the increase in employment was not too be attributed to another (state-related) cause. Neumark and Wascher (2006, p.18) point out that sometimes the effects might take some time and that this was not considered by Card and Krueger. Therefore, they question the strength of the conclusions drawn from this research. Though most research for minimum wage focused on the restaurant industry, Addisson, Blackburn and Cotti (2009, p. 5), focused on the retail industry, but they found a slight rise in employment after a rise in minimum wage as well. Furthermore, they examined nationwide data, eliminating possible interstate differences. Though an explanation of the rise in employment could be that hours are decreased, for which they have not tested, they deem this as unlikely, since they found that weekly earnings went up.

3.2.b Empirical findings in Europe

Surprisingly, there is little cross-country evidence available, in contrast to the abundant literature of country specific effects (Addison and Otzturk, 2012, p. 779).

As Hamermesh (2002) argues, international data are particularly valuable to minimum wage research, as long as international variations are taken into account. After discussing the cross country research, evidence found in individual European countries will be discussed.

Dolado et al. (1996) researched the effects of minimum wage in 4 countries in Europe; the

Netherlands, Spain, France and the United Kingdom. They noted that due to a higher minimum wage in Europe, it is tempting to blame that as a reason for high unemployment (p. 354). Finding very mixed results, they conclude that the importance of minimum wage has probably been exaggerated (p. 357). They argue that it is faulty to blame the increase in unemployment in Europe on the minimum wage. However, Neumark and Wascher (2004) did a cross-sectional analysis on 17 countries, though a lot of them non-European, concluding that there is enough evidence to indicate a negative effect on youth employment, controlling for country specific effects. They did find that the effects in the different countries fluctuated, according to them this could be attributed to two main areas. The first factor that influenced the impact of a minimum wage was whether this

minimum wage was set by collective bargaining, the second factor was the presence of labor market policies. According to the authors, such policies have a major influence on the impact of minimum wage. They found that the effects of minimum wage on unemployment where the strongest in least

(10)

9 regulated markets. Motivated by this study, Addisson and Otzturk (2012) did a cross-country

analysis, in a similar way, focusing specifically on the effects of a minimum wage on the female unemployment rate. They found that the negative impacts of a minimum wage where the strongest in countries with a big wage gap and high fertility rates.

Abowd, Kramarz, Lemieux and Margolis (1997) did a cross country analysis, comparing France with the United States. Controlling for variations between the two countries, they concluded that both in France as in the United States, there was a negative impact of a minimum wage. They also found that in both countries, youth employment had the strongest negative reaction.

Pereira (2003) researched the effects of the rise of an minimum wage in Portugal in 1987. She focused on the teenagers, whose wage had been raised significantly more than those of other age groups, creating a very natural experiment. They found that the raise significantly decreased youth employment, but raised employment of young adults (20-25), suggesting a substitution effect of labor. The same negative effects on youth unemployment were found in France, by Bazen and Skourias (1997, p. 730). Like Pereira (2003) and Card and Krueger (1992), the authors examined the effects of minimum wage in a natural experiment. They researched the effects of the wage increase in 1981, when the rise in minimum wage was significantly larger than in previous years (10%

compared to the usual 2%). They found that youth unemployment increased as a result.

According to Neumark and Wascher (2006, p. 88), the effect of minimum wage on unemployment has been studied in a more extensive way in the UK than anywhere else, besides of the United States. They state that the research done in the United Kingdom is widely used to prove that there are no significant negative effects that minimum wage reduces employment, and the research there is therefore particularly significant. The United Kingdom Machin, Manning and Rahman (2002, pp. 154-180) looked at the effects of the minimum wage in the United Kingdom, after the

implementation of a national minimum wage in 1999. They focused on studying the effects of the minimum wage on UK residential care homes industry, which can be viewed as being extra vulnerable to minimum wage, according to the authors. The authors found reduction in both employment and hours. However, as they do point out themselves, one should be careful with extrapolating their results to more general conclusions about the economy, since they are focusing on a sector where minimum wage has a substantial impact. Neumark and Wascher (2006, p. 82) are critical of their research, stating that the analysis is too simple, and ignores other changes that may have influenced this sector. Stewart (2004) also examined the effects on a minimum wage on the entire UK in the second era, examining the most recent effects of the minimum wage

(11)

10 than the minimum wage. The author also checked for four demographic groups (male, female, youth and adults). However, he found no significant change in employment due to the implementation of minimum wage in any of the demographic groups. Neumark and Wascher (2006, p. 85) are critical of these findings, stating that Stewart, as well as Card and Krueger, did not consider the fact that the effects might be lagged. The authors further stress that the question about long-run effects have yet to be addressed in the United Kingdom.

Though there is some evidence that a minimum wage might have negative effects on unemployment in the United States, there is enough evidence to suggest the contrary. Economists are heavily divided on this topic, in the United States, as well as in Europe. However, Neumark and Wascher (2006, p. 104) did an extensive literature review, comparing the wide range of research there has been done on this subject, in many countries, such as the United States, the United Kingdom, Spain, the Netherlands, Australia, Portugal and Greece. They conclude that is it false to assume that the most recent research does not support the classical view. They point out multiple flaws in the studies that fail to find a negative effect. The authors state that, as well as not finding discordant results, the summary of the research is also presented differently. Some research is being concluded as ambiguous, others state that there is no effect, or even a positive effect. However, Neumark and Wascher state that in their own research, they tend to find negative employment effects in the lower-skilled groups (2006, p. 105). Also, they emphasize that lagged effects of minimum wage truly do matter and are not incorporated enough in some of the research which fails to find negative effects. They state that if studies would focus more on long-run effects, more evidence for the classical view would be found (2006, p. 121).

(12)

11

4. Econometric Analysis

4.1 Methodology and Data

In this research I will do a pooled regression, focusing on countries with a national, statute minimum wage, so not set on bargaining agreements, keeping more factors consistent. To control as much as possible for differences between countries, since I am doing a pooled regression, only West

European countries, that are a member of the European Union, have been chosen, since there was more data available.

The countries I will use are:

Luxemburg Portugal France Ireland Spain The Netherlands The United Kingdom Belgium

The United Kingdom only has a national minimum wage since April 1999, therefore I will use data from 2000-2014. All the data is collected from the website of the OECD, the Organization for Economic Co-operation and Development.

The regressions I will be doing are based on the one used by Neumark and Wascher (2004): 𝑈𝑈 = 𝛽𝛽1 + 𝛽𝛽2 𝑀𝑀𝑀𝑀 + 𝛽𝛽3 𝑇𝑇𝑈𝑈 + 𝛽𝛽4𝑆𝑆𝑈𝑈 + 𝛽𝛽5 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺ℎ + 𝛽𝛽6 𝐺𝐺𝐺𝐺𝐺𝐺𝑆𝑆𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 + 𝜀𝜀

𝑌𝑌𝑈𝑈 = 𝛽𝛽1 + 𝛽𝛽2 𝑀𝑀𝑀𝑀 + 𝛽𝛽3 𝑇𝑇𝑈𝑈 + 𝛽𝛽4𝑆𝑆𝑈𝑈 + 𝛽𝛽5 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺ℎ + 𝛽𝛽6 𝐺𝐺𝐺𝐺𝐺𝐺𝑆𝑆𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 + 𝜀𝜀 U= unemployment of the entire population, defined as the percentage of the total labor force.

YU= represents the youth unemployment as a percentage of the total youth labor force, where the youth group is defined as the group aged 15-24.

MW = The minimum wage measure will be defined as a ratio to the median wage, as computed by the OECD. Neumark and Wascher (2004) use the ratio to the average wage, stating that this is the standard indicator used in the literature on minimum wage, and captures the effects of the minimum wage on the wage distribution. However, the OECD

(13)

12 contest that the median is a better basis for international comparisons, because it accounts for difference in earnings dispersion across countries, such as bonuses. I will therefore use the ratio computed on the basis of the median wage.

TU= Trade Union Density. Trade union density corresponds to the ratio of wage and salary earners that are trade union members, divided by the total number of wage and salary earners. Like Neumark and Wascher (2004), I included this variable in my model, since I suspect that trade union density affects employment in a positive way, it might be harder to fire employees in a country where a lot of workers are member of a union, and have a higher bargaining power.

SU = Strictness of Unemployment. This indicator includes all the procedures and costs involved in dismissing/hiring employees, individuals as well as groups of workers. Those indicators have been compiled by the OECD using statutory laws, collective bargaining agreements, case laws, contributions from OECD officials as well as advice from country experts. This variable is included, because employment might very well be affected by how difficult hiring/firing is, as contested by Neumark and Wascher (2004). I suspect that countries with a higher strictness of employment will have lower unemployment rates, since it is harder for employers to fire employees. Employees will not resort to firing as quickly.

GDPgrowth = Where GDP in Us Dollars per capita, and the growth is a percentage per year. I included this variable to control for effects due to general economic conditions, as also done by Brown (1982). I expect GDPgrowth to have a positive influence on employment.

Govspend = percentage of the GDP spent on unemployment by the government. The OECD defines it as all the expenses for people to compensate for unemployment. This is used as a measure as to how lenient countries are to the unemployed. Neumark and Wascher (2004) include a more elaborate variable in their model, including all the spending of the

Government on unemployment, including their investments to boost unemployment. However, due to lack of sufficient data, I included this variable instead. Neumark and Wascher (2004) point out that a countries which invests more in unemployment through different policies, might have lower unemployment rates. However, now I expect the Government Spending to have a negative impact on unemployment, since it is probable that countries, in which the unemployed are given a lot of compensation, will be less motivated to find a (new) job very quickly, thus resulting in higher unemployment.

(14)

13 0 5 10 15 20 25 U ne m pl oy m e nt 2000 2005 2010 2015 Year BEL ESP FRA GBR IRL LUX NLD PRT

My estimation method will be OLS. I will use the Breusch-Pagan test to check for heteroscedasticity, if this is indeed the case (α=5%), also a regression using robust standard errors will be done.

4.2 Hypotheses

The main goal of this thesis is to see if there is a relationship between minimum wage and

unemployment, as well as between minimum wage and youth unemployment. I do not expect the minimum wage to have a (negative) impact on unemployment, mainly because the stronger, more elaborate studies done in Europe have yet failed to find such an effect.

Hypothesis 1: There will be no significant impact of minimum wage on unemployment.

There are, however, more indicators that a minimum wage could have more effect on the youth unemployment from previous studies. So I expect there to be a rise in youth unemployment due to an increase in minimum wage, but no significant effect on general unemployment.

Hypothesis 2: There will be a significant impact of minimum wage on youth unemployment.

5. Results

5.1 Results Unemployment

The unemployment in Western Europe has developed as follows in the last few years:

Figure 3

(15)

14 Before starting with the regressions, a summary of the data:

Table 1: Summary Unemployment

This figure shows a summary of unemployment, 2000-2014.

Country N Mean Std. Dev. Min Max

All Countries 124 8 4.4 2.25 26.09

Spain 15 15.32 6.4 8.23 26.09

We immediately see that the Spanish unemployment is much higher than the other countries. Therefore, besides the first regression, I will also perform a Chow Test for structural break,

comparing the complete model with all the data to a regression omitting the Spanish data, since the Spanish data might influence the data in a way that does not fit the model.

The first OLS regressions with robust standard errors gives the following estimates:

Table 2: Regression of Unemployment, using robust standard errors.

The regression is based on annual data of 2000 till 2014. The T-values are included between brackets under the coefficients.

Dependent Variable: Unemployment

Variable Coefficient P-Value

Constant 15.39 0.000

Minimum/Median wage ratio

(6.6) -12.066

(-3.03) 0.003

Trade Union Density -0.135

(-7.63) 0.000 Strictness of Unemployment -1.026 (-3.08) 0.003 GDP Growth -0.142 (-1.37) 0.018 Government Spending 2.94 (8.86) 0.000 Breusch-Pagan Test 0.0106

(16)

15 All the coefficients are significant at a 5% level. It is remarkable however, that the coefficient of the minimum/median wage ratio is negative. This implies that a rise in minimum wage would actually lead to a decrease in unemployment. Those results were also found by Pereira (2003), and would indicate that there might be more of a monopsony. GDP growth, Trade Union Density and Strictness of Unemployment indeed have a negative on unemployment (and thus a positive one on

employment). Government Spending indeed has a negative effect on employment.

The second regression, omitting the Spanish data, gives the following results:

Table 3: Regression of Unemployment, omitting the Spanish data, using robust standard errors. The regression is based on annual data of 2000 till 2014. The T-values are included between brackets under the coefficients.

Dependent Variable: Unemployment

Constant 6.25 0.121

(1.57) 3.44

(0.55) 0.587

Chow-Test for structural break 0.000

From the Chow-Test, we see that the hypothesis that the Spanish data can be modelled the same way as the data from the other countries is rejected. Examining the minimum wage coefficient, we see that it is now positive, though not significant. Though all the coefficients got higher p-values, and shifted a little, the difference between the minimum/median wage ratio coefficients is very

(17)

16 0 20 40 60 Y o ut h Un em p lo y m e nt 2000 2005 2010 2015 Year BEL ESP FRA GBR IRL LUX NLD PRT

minimum wage affects unemployment. The control variables do not show any significant change after omitting Spain, regarding their effect (positive/negative).

Though in the model with Spain I found a positive impact of minimum wage on the unemployment, this effect was not there anymore when Spain was omitted from the regression. However, there is not enough statistical evidence to suggest that a (rise in) minimum wage increases unemployment.

5.2 Results youth Unemployment

The development of youth unemployment:

Figure 4 : Development Youth Unemployment.

(18)

17 Again, a short summary of the data:

Table 4: Summary Youth Unemployment

This figure shows a summary of Youth unemployment, 2000-2014.

Country N Mean Std. Dev. Min Max

All Countries 124 18.60 9.75 4.89 55.48

Spain 15 32 14.18 17.89 55.48

From the figure and table above, we see that youth unemployment is much higher than

unemployment, and again that the youth unemployment in Spain is much higher than the one in the other countries. So, like with the regressions above, I will do additional regressions, omitting Spain from the model, and performing a Chow test to see if there is a structural break.

The OLS regressions with robust standard errors gives the following estimates:

Table 5: Regression of Youth Unemployment, using robust standard errors.

Dependent Variable: Youth unemployment

Constant 24.77 0.000

Minimum/Median wage ratio (5.12) -8.31

(-1.07) 0.287

(19)

18 Though the minimum/median wage ratio now again has a negative coefficient, it is not significant. Again, GDP growth, Trade Union Density and Strictness of Unemployment indeed have a negative on youth unemployment (and thus a positive one on youth employment). Government Spending now even has a larger positive effect on youth unemployment. Testing the same model, but now by omitting Spain, I find the following results:

Table 6: Regression of Youth Unemployment, omitting the Spanish data

Dependent Variable: Youth Unemployment

Constant 3.81 0.559

(1.57) 28.19

(2.39) 0.019

Chow-Test for structural break 0.000

Again, it is remarkable how much the coefficient of the minimum/median wage ratio changes when Spain is omitted from the model. The coefficient now is very high, and now significant, suggesting that there is indeed a negative effect of a minimum wage on youth employment. Also, there is no longer any statistical evidence for heteroscedasticity, as there was in the model including Spain. Like with unemployment, the other coefficients have changed a little bit, but not as drastically as the minimum/median wage ratio, their sign staying unchanged.

So, when Spain was included in the model, minimum wage had a statistically significant effect on unemployment, but not on youth unemployment. By omitting Spain from the model, the effect of minimum wage on unemployment is no longer significant, however, the effect on youth

(20)

19

6. Conclusion and Discussion

6.1 Conclusion

Ample research has already been done on the effects of minimum wage on unemployment, most of in the United States. Not only theories, but also empirical evidence seem unable to find a consensus in the matter. Though, in Europe specifically, there has not yet been any statistical significant proof that an increase in minimum wage negatively impacts employment. In this thesis a pooled analysis was done in 8 European countries, looking at both the effects of a minimum wage on

unemployment, as well as on youth unemployment, using OLS.

Both a regression with Spain and without Spain has been done. By performing a Chow test I concluded it is better to omit the Spanish data from the model. No statistical evidence has been found to suggest that a minimum wage has a negative impact on employment. However, after omitting Spain from the model, at 5% significance, it can be concluded that a minimum wage has a negative impact on youth unemployment. After omitting Spain from the model, both my hypotheses turned out to be correct.

6.2 Discussion

Like Dolado et al. (1996), I found very mixed results. The first regression showed that a minimum wage even had a statistically significant positive influence on unemployment, like also found by Pereira (2003), but after omitting Spain from the model, this positive result was gone, and there was no significant result anymore, as also shown by Stewart (2004). The results for youth unemployment were statistically insignificant at first, but after omitting Spain from the model, a statistically

significant effect was seen, suggesting that a minimum wage doe have a negative impact on youth unemployment, like research done by Bazen and Skourias (1997, p. 730), Pereira (2003) and Zavodny (2000).

6.3 Limitations and Further Research

The sample used now consists of only 8 countries, future research might be advisable, with more countries, especially since so little cross country studies have been done. Also, panel research should be done, controlling for fixed country effects, but it did not fit in the scope of this thesis.

(21)

20 Limited data is always a problem in research. As an indicator of how countries use active labor market tools to boost unemployment, government spending on unemployment has been used. In future research, more control variables should be used for this, since there are more ways the government can boost employment.

Furthermore, more factors should be controlled for then, since the more countries you add, the more differences between countries there will be, which need to be accounted for in the

econometric model. In a few years, when the effects of the minimum wage in Germany are more clear, Germany could be added to this same regression, adding more data, not only because of an extra country, but also because of more years with information.

Additionally, as stated in the literature review, it would be useful and prove very insightful to do more research as to whether the low-wage employment market behaves according to the neo-classical model, or is better explained by a monopsonic model.

(22)

21

References

Abowd, J. M., Kramarz, F., Lemieux, T., & Margolis, D. N. (1997). Minimum wages and youth

employment in France and the United States (No. w6111). National Bureau of Economic

Research.

Addison, J. T., Blackburn, M. L., & Cotti, C. D. (2009). Do minimum wages raise employment? Evidence from the US retail-trade sector. Labour Economics, 16(4), 397-408.

Addison, J. T., & Ozturk, O. D. (2012). Minimum wages, labor market institutions, and female employment: A cross-country analysis. Industrial & Labor Relations Review, 65(4), 779-809. Bazen, S., & Skourias, N. (1997). Is there a negative effect of minimum wages on youth employment

in France?. European Economic Review, 41(3), 723-732.

Brown, C. C., Gilroy, C., & Kohen, A. I. (1982). The effect of the minimum wage on employment and unemployment: a survey.

Card, D., & Krueger, A. B. (2000). Minimum wages and employment: a case study of the fast-food industry in New Jersey and Pennsylvania: reply.American Economic Review, 1397-1420. Dickens, R., Machin, S., & Manning, A. (1999). The effects of minimum wages on employment:

Theory and evidence from Britain. Journal of Labor Economics, 17(1), 1-22.

Dolado, J., Kramarz, F., Machin, S., Manning, A., Margolis, D., & Teulings, C. (1996). The economic impact of minimum wages in Europe. Economic policy, 11 (23), 319-372.

Organisation for Economic Co-operation and Development. Data retrieved from: http://stats.oecd.org/

Hanke, S. (2014). Let the Data Speak: The truth behind Minimum Wage Laws. Retrieved from: http://www.cato.org/publications/commentary/let-data-speak-truth-behind-minimum-wage-laws

Hamermesh, D. S. (2002). International labor economics. Journal of Labor Economics, 20 (4), 709-732.

Lang, K., & Kahn, S. (1998). The effect of minimum-wage laws on the distribution of employment: theory and evidence. Journal of Public Economics, 69(1), 67-82.

(23)

22 Neumark, D., & Wascher, W. (2004). Minimum wages, labor market institutions, and youth

employment: a cross-national analysis. Industrial & labor relations review, 57(2),

d 223-248.

Neumark, D., & Wascher, W. (2006). Minimum wages and employment: A review of evidence from the new minimum wage research (No. w12663). National Bureau of Economic

Research.

Machin, S., Manning, A., Rahman, L. (2002). Where the minimum wage bites hard: Introduction of minimum wages to a low wage sector. Journal of the European Economic Association, 1 (1), 154-180.

Pereira, S. C. (2003). The impact of minimum wages on youth employment in Portugal. European

Economic Review, 47(2), 229-244.

Pindyck, R., & Rubinfeld, D. (2009). Microeconomics. Pearson: London

Slonimczyk, F., & Skott, P. (2012). Employment and distribution effects of the minimum wage.

Journal of Economic Behavior & Organization, 84(1), 245-264.

Stewart, M. B. (2004). The impact of the introduction of the UK minimum wage on the employment probabilities of low-wage workers. Journal of the European Economic Association, 67-97.

Zavodny, M. (2000). The effect of the minimum wage on employment and hours. Labour

The effect of minimum wage on unemployment in Western Europe