The youth employment rate and the role of minimum wage, labor market policies and institutions: theory and estimates

Abstract

Growth and labour market forecast in OECD countries remains bleak for 2013, especially for youth. The modest employment recovery in 2011 did not continue in 2012: youth unemployment rates stagnated at 15.7% in 2014 (OECD: Employment Outlook, 2013). This is well above pre-crisis rates, 11.8% in 2007. According to the OECD, the modest increase in GDP forecasted for 2013 (1.2%) and the subsequent slow recovery in 2014 (2.3%) will be insufficient to create employment and reduce unemployment in a substantial way.It is particularly alarming that one every 11 young people are now unemployed, not in full time studies or in employment. Also youth have lower access to unemployment benefits that adults, for lack of employment history (OECD, 2010, off to a Good Start, Jobs for Youth Synthesis Report). These trends have contributed to higher rates of poverty and social exclusion and increasing polarization in society and in the labour market. For instance, living standards decreased in 15 Member States in 2010 with respect to the year before (Eurostat, March 2013). These issues represent tremendous social and economic costs to society, in terms of worsening social dislocation, skills loss, violence and crime, as much as they denote important failures to get unemployed back to productive and sustainable jobs, and to protect youth at risk of poverty. Aggregate demand is low and this certainly explains high levels of youth unemployment rates. However a number of barriers may prevent that demand translate into higher employment for youth. In particular, vocational training might not be well adapted to changing labour market requirements. In addition, there could be demand-side obstacles to youth employment. Employers, for example, might be faced with high social security contributions for low-paid work, or face high minimum wages. There might be insufficient support to help the young unemployed to find work. Finally, jobs prospects for youth are hampered by limited regional mobility in some countries of Central and Eastern Europe and South Europe (OECD, Jobs for Youth, Synthesis Report, 2010). This thesis seeks to measure whether high levels of minimum wages could explain rising unemployment rates for youth in OECD countries, other things equal. It also looks at whether other labour market institutions, such as strict Employment Protection Legislation (EPL) can explain low employment levels (for youth). It takes into account also the role paid by active labour market policy, collective agreement and a youth sub-minimum wage for 2 youth. The analysis uses a cross-section of panel data on minimum wages over the period 2000 to 2011, while from 2000 to 2008 for others employment protection legislation, and from 2004 to 2011 for active labour market policies indictors, including 22 OECD. Following Bassanini and Duval(2010), a panel data model has been used including GMM indicator, using the same data base but including different time period. The main conclusions are that, minimum wage, measured with Kaitz Index, has a negative impact on youth employment. Additionally, some ALMP’S seem to show a positive effect on youth employment(elasticity); this is a very important thing mainly because it has never been estimated in previous paper(only the theory was able to support it). Considering labor market institutions, their impact depends by which one we consider: union density confirms its negative impact on youth employment rate, while on the other hand EPL variable using OECD definition confirms its positive effect on youth employment. ALMP’S have a great influence on youth employment rate, several variables are positive and significant, while other variable included in the model only in part confirms previous literature. The last part of the second chapter ends with a little discussion about gender discrimination in the issue of youth employment too. As results show, also if the sign of the variable is always the same (confirming both literature and the goodness of the model), however the magnitude tends to be stronger (in negative meaning), for female estimates... [edited by Author]