Vicent Almenar-Llongo (p), Celia Melguizo (p), Raúl Ramos

In the Spanish economy, economic fluctuations have had a great impact on the labour market and especially in unemployment. During the recent economic recession that took place between 2008 and 2013, the reduction in GDP in more than 16 percentage points resulted in an 18 percentage points unprecedented increase in unemployment rate, which reached 26 percent of the active population in 2013. The subsequent period of recovery also implied a strong variation in the unemployment rate that accounted in almost 9 percentage points. But, the Spanish territory has not been affected in a homogeneous way and a different unemployment sensitivity has been found (Villaverde and Maza, 2007, 2009; Melguizo, 2017). This work aims to contribute to previous literature by determining the effect of spatial spillovers in the relationship between GDP and unemployment rate for the Spanish provinces (NUTS-3 regions) for the period that ranges from 1985 to 2015. In order to do so, we resort to an Okun’s law extended version. But, first of all, we check the stationarity of the series and the panel of provinces by using the traditional individual and panel unit root tests. Then, we perform the Spatial Autoregressive Model (SAR), which only considers the endogenous variable spatially lagged but with non-spatial effects in the GDP variable. Later on, besides the spatial influence of unemployment rate of the other provinces, a spatial effect of the GDP variable is also included. Therefore, a Spatial Durbin Model (SDM) is applied. Preliminary results provide interesting findings. In the spatially lagged version, we obtain that regional GDP coefficients are significant for all provinces which showed high levels of unemployment sensitivity (β_1 ≤ - 0.3) in the baseline model (model without spatial effects). However, for provinces that showed a β_1 ≥ - 0.2 or a not significant coefficient in the baseline model, relationship between the aforementioned variables turns out insignificant and, also, for most provinces in which Okun’s relationship turns out insignificant, spatial effect has a significant influence.

Angel Martin-Roman (p), Alfonso Moral , Javier Martin-Roman (p)

The economic cost of workplace accidents is not taken into account in the social policy debate often. However, these costs are very high. Krueger and Meyer (2002) show that during the second half of the nineties the amount of the workers’ compensation was twice than unemployment benefits. In Spain, some reports of the most important Trade Unions also focus on this topic (CC.OO., 2004).
The analysis of regulation related to workplace accidents has a long tradition in North America. In 1970, the United States Congress passed the Occupational Safety and Health Act, which established the Occupational Safety and Health Administration (OSHA). In Spain, the Labour and Social Security Inspectorate (Inspección de Trabajo y Seguridad Social, ITSS) is a public service which is responsible for carrying out surveillance of the social rules compliance and for enforcing the responsibilities in case of offense or breach. The aim of this paper is to analyze whether the ITSS, as the organization responsible for enforcing labour rules in the prevention of occupational hazards, is effective in carrying out its purpose and reducing such occupational hazards.
In this study, we apply a spatial dynamic model based on Roodman’s proposal (Roodman, 2009) and use data from the Spanish Labour Force Survey, the Spanish Statistics on Workplace Accidents and the annual reports developed by the ITSS. The units of analysis are the 50 Spanish provinces (NUTS-3) and the period of time covered goes from 1994 to 2014. The results obtained do not show an unambiguous sign for the inspection variables. First, we detect a negative labour inspectorate effect that reduces the workplace accident rate, that is, the ITSS intervention could become an effective tool in reducing it. However, the number of inspectors presents the opposite effect, a fact that could be supported by a claim reporting effect: increased surveillance encourages workers to report some accidents that, otherwise, would not have registered. Additionally, we implement an analysis that highlights the relevance of the spatial dimension (Beloti et al., 2017; Shehata and Mickaiel 2013). Previously, Bande and López-Mourelo (2014) had found some evidence in this favor, but this work constitutes a step forward focusing on the development and potential of the most recent econometric techniques.

Miriam Marcen (p), Rafael González-Val

Spain is characterized as a country with pronounced recessions and significant volatility in unemployment rates. From 1998 to 2001, the average unemployment rate fell dramatically. Then, after a stable rate around 10% until 2004, there was detected another fall until 2007, but during the period known as the Great Recession the rate has followed a very steep slope, reaching levels of almost 25%.
In this paper, we examine the effect on regional employment of the Spanish labour law reform passed in 2012 to change substantial issues of the labour market. Among other things, the reform reduced dismissal cost and the bargaining power of labour unions.
We use data from 50 Spanish provinces (NUTS III regions) from 1985 to 2018, considering the registered number of employees and the employment and unemployment rates. By using panel data models with fixed effects, linear and quadratic region-specific time trends and other relevant control variables at the regional level, our results reveal that the reform had a significant effect on regional employment.

Albert Yirmiyahu (p), Ofir Rubin , Miki Malul

Most studies, which deal with the evaluation of large-scale reforms in education, typically focus on the likelihood of acquiring higher education and the associated returns in the labor market. Our previous research, which conducted on the basis of Central Bureau of Statistics (ICBS) data from 2008, indicates that such reform in Israel has made academic education accessible for the residents of the periphery and improved their earning ability. In this study we examine the dynamics of this improvements over time. Our empirical analysis uses data of individual income surveys for 2001-2016 in Israel (ICBS).

The estimation of wage equation as a function of the individual's characteristics generally yields a positive return to education, but taking the number of schooling years as an exogenous variable might cause bias in education coefficient. One of the reasons is that education is a choice variable correlated with unobservable attributes and characteristics such as inborn intelligence, acquired skills, and motivation. In order to overcome the correlation between level of schooling and various unobserved variables, we will focus on a particular shock in supply side of education. We will identify the effect of education on salary using observed variable (instrumental variable) that describes the policy change in the Israeli higher education system (i.e. Academic Colleges Law 1995) which has supported the accessibility of academic studies to a broad spectrum of groups and thereby has also contributed to reducing the wage gap in the long run.

The estimation of wage regression is processed by TSLS approach. Preliminary results indicate that the impact of the Academic Colleges Law on the earning capacity and employment of individuals in the labor market has diminishing marginal effect and stabilizes after several years. In addition, it was found that reform in higher education continues to contribute, but not to a large extent, to the narrowing of the wage gap between the two investigated groups in the Israeli labor market (core vs. periphery). These gaps seemingly stem from the quality of educational institutions and the fields of study chosen by students of periphery regions compared to residents of the core in Israel.