In recent decades, the Spanish labour market has been characterised by high levels of unemployment and temporary work. In this article, we analyse the intergenerational effects of temporality in the labour market and loss of work, focusing on children's education. Among the results that stand out, we find that both employment insecurity (measured by the type of work contract held by the father) and the father's loss of work, lead to a decline in the educational performance of children. However, we do not detect significant effects on children's educational results in our analysis of the same variables for the mother. Our findings suggest that it is important to take into consideration the intergenerational effects of employment policies when designing or reforming them.

In recent decades, the Spanish labour market has been characterised by high rates of unemployment and temporality. The latter phenomenon is the consequence of the reforms implemented in the 1980s to introduce flexibility in the labour market. Since then, the Spanish economy has been one of the best examples of the development of a dual labour market, that is, a labour market in which workers form part of two clearly differentiated groups. On the one hand, there are workers with indefinite contracts (also referred to as fixed or permanent) who enjoy a high level of protection (so-called insiders). On the other, there are workers with temporary contracts (outsiders) who are minimally protected from job loss.

Graph 1 shows the percentage of workers with temporary contracts in different European countries, as well as the average for the EU-15, based on the first year with available data. With values above 25% until the beginning of the crisis, Spain is the country with the highest rate of temporary employment, even after the significant destruction of temporary work that has occurred since the recent recession.

In addition to the characteristics of collective bargaining in Spain, the regulation of competition and the size of Spanish firms, the dual labour market is one of the reasons why Spain has one of the highest unemployment rates in Europe, currently only below Greece. Graph 2 shows that, starting in the second half of the 1980s, the unemployment rate declined from approximately 20% to 15% before the crisis of the 1990s. With that crisis, unemployment returned to over 20% and more than a decade would pass before it would reach levels similar to other European countries in 2007, just before the recent recession. Since then, the massive destruction of jobs has been one of the most important characteristics of the recent recession on both sides of the Atlantic. In the United States, the data reveal a record in terms of the destruction of employment, with an unemployment rate that reached 10% in the first quarter of 2010. In Europe (the EU-27), based on data from the Labour Force Survey, from 2008 to 2012 there was a net loss of 7 million jobs, with almost 60% of those jobs lost in Spain. In addition, since the beginning of the recent recession, Spain has been characterised by difficulties in finding employment, resulting in a rise in long-term unemployment.

In this article we analyse the intergenerational effects of the dual labour market and the loss of employment. In particular, we look at the impact that the father's employment contract (temporary versus permanent) has on the probability of children finishing secondary education on schedule. In the economic literature, based on our understanding, there are no studies that attempt to identify the causal impact of parents' work contracts on variables that measure the educational performance of their children.

In addition, we describe the effect of parents' loss of employment on the educational performance of their children during obligatory education. Other studies (see section 3) have attempted to examine similar issues for other countries, while this is the first study that uses Spanish data and that focuses on studying the impact of job loss during the recent recession. Before examining our results, we take a brief look at the literature that will help us understand why parents' temporary employment and unemployment affect the educational development of their children.

Various studies in the economic literature have analysed whether being employed with a temporary contract instead of a permanent one has adverse consequences for the worker. Bentolila et al (2008) and Dolado et al (2002) offer a review of the empirical literature that looks at these consequences. Based on the conclusions from these studies, workers with temporary contracts receive a lower investment in training on the part of employers and lower salaries for the same type of work. This wage gap between temporary and permanent workers is also associated with the fact that firms employ temporary workers in positions that require lower qualifications than they possess. Findings also show that being employed with a temporary contract is associated with a greater risk of suffering a work-place accident (Guadalupe, 2003) and lower job satisfaction (Booth et al, 2002).

Studies in the field of social psychology relate parental job insecurity with lower academic performance among children (for example, Barling et al 1999). Along with the economic literature just discussed, these studies suggest various mechanisms for why parents' work contracts could affect their children's educational results. First, several of the consequences mentioned above affect variables (income, job satisfaction, health, etc.) that have traditionally been seen as determinants of children's school performance. Secondly, as suggested by one of the branches of social psychology, fathers' job insecurity can negatively affect the effort and attitude of children toward schoolwork (Lim and Loo, 2003).

In this section we use data from the Spanish Labour Force Survey [Encuesta de Población Activa (EPA)] for the years 2000 through 2004, to determine if the type of employment contract held by fathers has an impact on the probability of their children finishing obligatory secondary education on schedule. The reader may ask why we focus on fathers' work contracts (and not mothers), as well as the reasons for choosing the probability of finishing secondary education on time as the main measure of the educational development of children. Regarding the first question, we focus primarily on fathers' contracts because the father continues to be the main source of household income. In addition, the psychological effects associated with job insecurity are more pronounced in men than in women (Lim and Sng, 2006).  In addition, we have looked at the impact of mothers' type of contract on the probability of graduating on time and we have not found any effect. The answer to the second question is determined by the availability of data.

The data from the Labour Force Survey allows us to study what impact the father's type of work contract (measured in the year in which the child should finish obligatory education) has on the probability of finishing obligatory secondary education on time (in other words, without repeating a school year). Graph 3 shows how much more probable it is that a student whose father is employed with a permanent contract finishes secondary education on time, compared to a student whose father is employed with a temporary contract. The estimate of these probabilities has taken into account a series of variables that can influence both the probability of having a temporary contract and the academic performance of children. That is, the results that are shown in the graph are not due to regional differences, nor to the educational level of the father or his employment experience, nor to the sector or occupation in which he is employed, among other relevant variables.

In line with what we find from the review of the literature in the beginning of this section, the first bar in graph 3 shows that it is more probable that students whose fathers are employed with a permanent contract will finish secondary school on time; in particular, they are 7 percentage points more likely to do so. Taking into account that 61% of the students in the sample whose fathers have permanent contracts finished obligatory secondary education at 16 years of age (on time), and that only 41% of those whose fathers have a temporary contract did so, these 7 percentage points explain 35% of the difference observed. 

The impact described is an average effect, that is, it is the average result obtained when we take into account the sample selected for this study. In what follows, we analyse if this average effect is different for specific sub-groups of the population. First, in the second and third bars in the graph we see the results depending on the maximum level of education reached by the father. They indicate that the effect is greatest for the sub-sample of students whose fathers have a higher level of education, although the difference is not statistically significant.

The last two bars in the graph show the differential impact of the father's type of contract on boys and girls. Although boys whose fathers have a permanent contract have a probability 9 percentage points higher of finishing obligatory secondary education on time than those whose fathers only have temporary contracts, for girls, the type of contract their fathers have does not seem to be important in explaining when they finish (the effect shown for girls is the only one in the graph that is not statistically significant). The social psychology literature tends to show that identification with parents acts as a mechanism that modifies the effect of the father's employment insecurity and the attitudes of children regarding work and effort (Barling et al, 1998). Thus, if boys identify with their fathers more than girls, they will be more affected by their father's employment situation, as the data shows.

Lastly, more technically advanced studies (which permit us to dig deeper in identifying causal effects) confirm our results, indicating that the advantage for students whose fathers have permanent work contracts during the year in which they should finish obligatory secondary education is even greater than shown in this section (Ruiz-Valenzuela, 2014, ch. 3).

The labour market, and in particular, the changing unemployment rate, have been constant sources of news on television and in newspapers since the beginning of the crisis in Spain. Similarly to what has been described in the previous section, various studies have analysed the consequences of job loss on the worker. Among other negative effects, individuals that lose their jobs suffer short-term lose of wages that seems to persist in the long-term (Jacobson et al, 1993), have a greater risk of divorce (Charles and Stephens, 2004), and suffer a decline in their physical and mental health (Eliason and Storrie, 2009). Closely related to this latter effect, a sharp increase in the number of mental disorders has been found among workers exposed to the shock of the construction industry in Spain (Farré et al., 2015).

With these results in mind, we have analysed the intergenerational impact of parental loss of employment (particularly that of the father) on variables that measure different aspects of children's academic performance. Among the most important results, we find a decline in students' grades at the end of obligatory secondary education (Rege et al, 2011), an increase in the probability of repeating a school year (Stevens and Schaller, 2011), and a lower probability of going on to university (Coelli, 2011).

In this section, we look at the intergenerational impact of shocks to the labour market, using data on the loss of parental employment and academic performance during the recent recession. This could be an additional cost of recessions that, up until now, has been undervalued in the academic literature.

To carry out our analysis we use data from 358 students who were in obligatory education (both primary and secondary) in 2012 in a school in the province of Barcelona. This school has characteristics similar to the average school in Catalonia (see Ruiz-Valenzuela, 2015). One of the advantages of this data is that it is panel type data; that is, we have data on students' educational performance and the employment situation of their fathers over various years (from academic year 2007-2008 to 2011-2012). This allows us to compare the academic performance (measured by students' overall average for all subjects) of the same student before and after the beginning of the crisis - in other words, before the father's loss of employment. Thus, this type of data reveals whether the effect we are measuring corresponds with the effect of the father's loss of employment or that of other variables, such as, for example, parents' level of education or their occupation.

Graph 4 shows, in the first bar, the decline in the average grade associated with father's loss of employment. The decline is an average of more than 0.2 points on a scale from 1 to 10, where passing is a grade of 5. The negative impact of a father's loss of employment during the recent recession in Spain, although it may seem minor, is double the effect of the loss of work found in other studies for other countries. In addition, bars two and three show that the effect is much greater for those students whose fathers lost their jobs during the crisis and had not found work after at least one year (i.e., those who suffer long-term unemployment). In this case, the average grade of students declines by almost one half point.

There are two additional observations we can make. First, boys suffer a greater decline in their grade average than girls. In fact, the slight decline that we see in the graph for girls is not statistically significant, that is, we cannot discard the possibility that there is no effect. The second has to do with the impact of the mother's loss of work on the academic performance of children. Just as when we looked at the effect of the mother's work contract, the loss of work of the mother does not appear to affect children's grade average.

Due to the high correlation between temporary work and exiting the labour market, the reader might think that the effect that we find here is more closely related to father's type of contract than to the loss of employment per se. However, it is important to emphasise that the majority of job loss found in our sample corresponds to employment of a permanent nature. In addition, the data indicate that the effects of the father's loss of work are even greater when the sample used includes only the loss of employment from a permanent job held before the beginning of the crisis. This is not surprising; the academic literature in the field of labour economics has found that the drop in income level, as well as the difficulty in finding employment after losing a job, are greater the more time spent working in the same job before losing it (see Jacobson et al, 1993).

The available data suggests that both fathers' temporary employment and loss of employment have negative effects on variables that measure the educational development of their children. In this study we have also found that there is no observable relationship between type of contract or the loss of employment on the part of mothers and the educational results of their children. This difference in the results found for fathers and mothers is consistent with results from studies in health economics and social psychology, which have documented more severe mental disorders and physical problems associated with unemployment and work insecurity for men than for women. Thus, some studies have shown, for example, that men suffer higher levels of stress and financial anxiety from both work insecurity and the loss of employment (Lim and Sng, 2006).

The findings of this study underline the need to combine policies of business flexibility, necessary so that firms can make adjustments at specific times of the economic cycle, with policies that mitigate the cost of these adjustments in terms of the wellbeing of workers and their families. This would involve, for example, putting into effect policies aimed at reducing the temporality of the labour market and giving greater protection to workers (and not necessarily to the specific job) in the case of layoffs. An example of this type of policy is found in the flexicurity model existing in Denmark, which combines levels of employment protection with generous subsidies for unemployment and well-designed active employment policies, based on providing training courses and elimination of unemployment benefits if they are not followed or if several work offers adequate to an individual's qualifications are rejected (Bentolil, 2010).

Given the high rates of temporary employment and unemployment in Spain, it is important to have more data to analyse in detail the mechanisms behind the results found in this article. In addition, if data permits, it is important to examine whether the harmful short-term effects that we have observed extend into the long-term. Understanding this is essential for anticipating potential increases in inequality in our society.

Jenifer Ruiz-Valenzuela, researcher

Centre for Economic Performance, London School of Economics

Barling, J., K.E. Dupre and C.G. Hepburn (1998): “Effects of parents’ job insecurity on children’s work beliefs and attitudes”, Journal of Applied Psychology, 83(1).

Barling, J., A. Zacharatos and C.G. Hepburn (1999): “Parents’ job insecurity effects children’s academic performance through cognitive difficulties”, Journal of Applied Psychology, 84(3).

Bentolila, S. (2010): “Flexiguridad ampliada”, Nada es gratis blog [http://nadaesgratis.es/bentolila/flexiguridad-ampliada].

Bentolila, S., J.J. Dolado and J.F. Jimeno (2008): “Two-tier employment protection reforms: the Spanish experience”, CESifo DICE Report, 4.

Booth, A.L., M. Francesconi and J. Frank (2002): “Temporary jobs: stepping stones or dead ends?”, The Economic Journal, 112(480).

Charles, K.K., and M. Stephens (2004): “Job displacement, disability, and divorce”, Journal of Labor Economics, 22(2).

Coelli, M.B. (2011): “Parental job loss and the education enrolment of youth”, Labour Economics, 18(1).

Dolado, J.J., C. Garcia-Serrano and J.F. Jimeno (2002): “Drawing lessons from the boom of temporary jobs in Spain”, The Economic Journal, 112(480).

Eliason, M., and D. Storrie (2009): “Job loss is bad for your health. Swedish evidence on cause specific hospitalization following involuntary job loss”, Social Science & Medicine, 68(8).

Farré, L., F. Fasani and H. Mueller (2015): “Feeling useless: the effect of unemployment on mental health in the Great Recession”, Barcelona GSE Working Paper Series, 838.

Guadalupe, M. (2003): “The hidden costs of fixed-term contracts: the impact on work accidents”, Labour Economics, 10(3).

Jacobson, L.S., R.J. Lalonde and D.G. Sullivan (1993): “Earnings losses of displaced workers”, The American Economic Review, 83(4).

Lim, V., and G. Loo (2003): “Effects of parental job insecurity and parenting behavior on youth’s self-efficacy and work attitudes”, Journal of Vocational Behavior, 63(1).

Lim, V., and Q. Sng (2006): “Does parental job insecurity matter? Money anxiety, money motives and work motivation”, Journal of Applied Psychology, 91(5).

Rege, M., K. Telle and M. Votruba (2011): “Parental job loss and children’s school performance”, The Review of Economic Studies, 78(4).

Ruiz-Valenzuela, J. (2015): “Job loss at home: children’s school performance during the Great Recession in Spain”, CEP Discussion Paper 1364, London School of Economics.

Ruiz-Valenzuela, J. (2014): Job loss at home: children’s grades during the Great Recession in Spain. Methods. Essays on parental labor market characteristics and the academic outcomes of their offspring. EUI PhD theses; Department of Economics.

Stevens, A.H., and J. Schaller (2011): “Short-run effects of parental job loss on children’s academic achievement”, Economics of Education Review, 30(2).