English (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mail
Cited by SciELO 
Cited by Google 
Similars in
SciELO 
Similars in Google 
Permalink
Introduction
The debate on the educational and labor issues faced by the youth population continues being part of the international, regional and national agenda. Despite the progress observed in recent years, young people are a vulnerable group with significant barriers to achieving decent work (International Labor Organization (ILO), 2015; 2017). According to the Permanent Household Survey data of 2016 (4th quarter), only 10.2% of the population aged 18 to 24 years old in Argentina gets a job without deficits of quality of employment.
Some of these barriers are associated with the low educational level of young people when entering the labor market. Nearly 40% did not finish high school, and 22% left secondary without completing it. There is also an inverse relationship between the school performance of young people and early entry in the labor market. In addition, young employed exhibit a high level of informality and job instability (Bertranou & Casanova, 2015; Bertranou, Jiménez & Jiménez, 2018).
In Argentina, during the 2000s, the national government implemented a set of education and employment policies targeting this population group to address their educational and occupational vulnerability. Some of the measures were implemented to provide young people with better tools to make a successful labor transition, including their formalization at work. Thus, in early 2014 government created the "Programa de Apoyo a los Estudiantes Argentinos" -PROGRESAR- so that young people aged 18-24 can start, return or finish to compulsory schooling or higher education. Likewise, this program also offers young people training experiences and qualifying practices in work environments. Given the program access requirements, the potential coverage of PROGRESAR has increased substantially in recent years, placing it within active policies for young people with one of the highest coverage rates (Bertranou, Jiménez & Jiménez, 2018).
This document will analyze the impact of PROGRESAR on the employment situation of young people aged 18-24. In particular, our principal interest is analyzing their probability of access jobs of different quality. The decent work approach allows us to consider different aspects of job quality, not only those related to social protection but also to labor rights, employment opportunities and social dialogue (ILO, 2002). Various theoretical models expose diverse reasons to link educational and labor programs with labor transitions (Bosch & Campos-Vázquez, 2014; Bosch & Manacorda, 2012).
The evidence on evaluations of the impact of social and labor policies targeting young people is scarce in our country (Elias et al., 2004; Ministry of Labour, Employment and Social Security (MTEySS) et al., 2010; Alzúa, Cruces & López, 2016; Banco de Desarrollo de América Latina (CAF), 2015). In particular, the impact evaluations of PROGRESAR are practically non-existent. Perhaps this is due to its relatively recent implementation and the lack of available information to monitoring of the results obtained by the beneficiaries. An exception is a study by Di Giovambattista, Gallo & Panigo (2014) that developed an ex-ante evaluation or microsimulation to examine the possible result of the program on income distribution. The results indicate that, depending on the level of final adhesion to the program, PROGRESAR could reduce income inequality by up to 32% among the youth population. Also, this impact might be higher in the poorest regions of the country (Bertranou, Jiménez & Jiménez, 2018).
This study contributes to the existing literature in several aspects. So far, there are no similar studies that carry out an ex-post impact evaluation of the PROGRESAR in Argentina. Neither do any of the studies mentioned estimate the consequences of this program on the labor trajectories of young people, regarding both exit and entry into the labor market, nor analyze the impacts on the transitions between jobs with special emphasis on the quality of labor insertion. On the contrary, most studies on impact evaluation in Argentina concentrate mainly on analyzing non-dynamic outcome variables. Finally, we implement the decent work approach defined by the ILO (2002) to consider different attributes that define job quality beyond social security coverage.
Related literature
Impact evaluations of labor policies for young people are quite numerous, especially in developed countries. The results that arise from the meta-analysis developed by Card, Kluve & Weber (2010; 2017), based on a broad set of estimates for European countries, Canada and the United States, suggest that job-search assistance programs have relatively favorable short-term impacts, while on-the-job training programs tend to show better results in the medium term than in the short-run. Likewise, they observed that programs for young people are less likely to yield positive impacts than untargeted programs.
In Latin America, there is a growing literature analyzing the outcome of active labor market policies on a variety of labor indicators. The results obtained by Escudero et al. (2018), based on a meta-analysis of 256 impact estimates derived from 51 studies for Latin America, show that training programs are slightly more effective than other types of interventions -such as public works, employment subsidies, self-employment and micro-enterprise creation programs and labor market intermediation services-. Also, formal employment is the outcome category that is most likely to be impacted positively by these programs. Likewise, labor policies focused on young people show better results than those targetting older workers.
In relation to labor programs for young people in Latin America, the meta-analysis developed by Vezza (2014) shows that its effectiveness, from the point of view of the results obtained by the beneficiaries in the labor market, seems to be more correlated to its focalization and way of implementation than to the type of intervention. Comprehensive initiatives combining different interventions also show a better performance than programs with a single component (Vezza, 2014).1
The studies most related to this research are those that analyze the impact of conditional transfer programs targeted young population on their work trajectories and the quality of employment. Technical and professional training is one of the principal services offered to young people who are the target of this type of initiative. This is observed not only in the PROGRESAR, but also in the programs of several other Latin American countries such as the Jóvenes en acción program in Colombia (ECLAC and ILO, 2014). This program, targeting people between 16 and 24, was implemented between 2001 and 2005 and consisted of a monthly monetary transfer conditional on attending training classes or work internships. Attanasio, Kugler & Meghir (2011; 2017) develop an impact evaluation of Jóvenes en acción based on an experimental design. The results obtained by the authors show the positive influence of the program, between young beneficiaries, on the probability of working in the formal sector. This is observed in both the short and the long term. Likewise, the Plan Sectorial de Calificación Profesional -PlanSeq- and the Programa Nacional de Acceso a la Educación Técnica y al Empleo -Pronatec-, complementary to the Bolsa Família in Brazil, are further examples of training services targeting young people in the framework of a program of conditional transfers. Based on the propensity score matching method, Petterini (2011) developed an impact evaluation of the PlanSeq. The results suggest that participants are 19.6% more likely to find a job than those who did not participate. The Avancemos program in Costa Rica also consists of conditional cash transfer targeting families with adolescents and young people between 12 and 25. In this case, the program requires enrollment in any of the secondary education modalities of the formal education system. Mata & Hernández (2015) measure the impact of the Avancemos program on the student desertion, from quasi-experimental methods -propensity score matching and difference in differences-, and find a positive influence on both dropout and reintegration rates.
On the other hand, there are several studies analyzing specifically the impact of on-the-job training programs targeted young people, such as those about the Juventud y Empleo2 program in the Dominican Republic (Card, Kluve & Weber, 2010; Ibarrarán et al., 2019). The results of the evaluation carried out by Card, Kluve & Weber (2010) indicate a positive, albeit modest, influence on income for those who had a job. Similarly, Ibarrarán et al. (2019) point out significant impacts on labor formality, particularly for males. The long-term analysis of the authors indicates that these results are maintained and grow over time. On the other hand, Ñopo, Saavedra-Chanduví & Robles (2007) and Díaz & Rosas (2016) evaluate the ProJoven, the Peruvian youth labor training program, which provides beneficiaries with basic training of three months in the classroom and internships in low-skilled occupations. An interesting aspect of the design of this program is it promotes gender equality by encouraging female participation in occupations traditionally dominated by men and by providing subsidies so mothers can participate (Ñopo, Saavedra-Chanduví & Robles 2007). Eighteen months after participating in the program, the female employment rate increased approximately 15%, and women's labor income grew by 93%. Unlike the previous evaluation, that developed by Díaz & Rosas (2016) arises from one experimental design and examines the long-term consequences of ProJoven. The estimates obtained from this design show a positive influence on formal employment that varies according to the gender and age of the beneficiaries. Also, Acero et al. (2009) examine the impact of the Jóvenes Bicentenario program in Chile3 and observe a statistically significant impact in decreasing work inactivity and increasing employment.
The evidence on evaluations of the impact of social and labor policies targeted young people is scant in Argentina. The study of Elias et al. (2004) evaluates the impact of the Entrenamiento Juvenil program, which was implemented between 1993 and 1999 in Argentina, using non-matching and matching techniques based on propensity score estimate. This program offered courses to young people that were not employed and had a low level of education. The main consequence of the program was on wages, not on the possibility of finding employment. The MTEySS et al. (2010) reports assesses the Jóvenes con Futuro program to improve the employability conditions of young people. The results indicate that the program training courses favor inserting quality labor, returning to the educational system and finishing secondary school. According to this, the authors conclude that obtaining formal educational credentials and access to a first registered and stable job are significant factors that prevent the return to informal work. Alzúa et al. (2016), meanwhile, evaluate the impact of the Entra21 program in Córdoba. This initiative included life-skills, vocational training and internships with private sector employers. The results, obtained from an experimental design, indicate gains -of 8 percentage points- in formal employment in the short term, although these results tend to dissipate in the medium term. CAF (2015) reports that the Primer Paso program generates a significant increase in the probability of entering formal employment after the beneficiaries leave the program. This is an on-the-job training and internship program to increase employment in a formally registered firm, which includes a salary subsidy and has the peculiarity of being financed by a subnational government -province of Córdoba-.
Moreover, the quantitative impact evaluations of PROGRESAR are practically non-existent. Perhaps this is due to its relatively recent implementation and the lack of available information to monitor the results obtained by the beneficiaries. An exception is a study by Di Giovambattista, Gallo & Panigo (2014) that presents an ex-ante evaluation or microsimulation to examine the possible influence of the program on income distribution. The results indicate that, depending on the level of final adhesion to the program, PROGRESAR could reduce income inequality by up to 32% among the youth population. In addition, this effect would be larger in the poorest regions of the country (Bertranou, Jiménez & Jiménez, 2018). Some studies examine the characteristics and design of the program (Bertranou, Jiménez & Jiménez, 2018; Otero & Corica, 2017, De Avila, 2016; Gluz & Moyano, 2016; Bertranou & Casanova, 2015; D'Alessandre & Duer, 2015; Gandini, Maldonado & Moreno-Yunis, 2014; Mazzola, 2014; Marzonetto & Aguirre, 2014) or made a qualitative analysis of the PROGRESAR based on case studies (Lara & Salazar; 2017, Roberti, 2018). However, none of the previous investigations developed a quantitative evaluation of the impact of the program on the eligible youth group or beneficiaries.
Description of PROGRESAR
Decree No. 84/2014 created PROGRESAR in January 2014, including components of economic security and improvement of employability. Its general objective is to generate opportunities for social and labor inclusion through integrated actions that enable young people to improve their employability. For this, the program offers them the possibility of completing compulsory schooling, initiating or continuing higher education and carrying out training or practical experiences in work environments. The program is targeted to young people between 18 and 24 who belong to vulnerable social groups. Specifically, to access the program, young people must be unemployed or work informally or formally, and their personal and family income must not exceed a certain threshold.
The main benefit of PROGRESAR is a monetary aid4 conditional upon enrollment and attendance at an educational establishment to initiate or continue studies at any educational level accredited by the Ministry of Education or the Ministry of Labour, Employment and Social Security -MTEySS-. The program also offers assistance for professional training, guidance and labor intermediation through the MTEySS and care of dependent children, through the Ministry of Social Development when necessary. The access and permanence to this benefit are subject to certain requirements related to the fulfillment of the educational objectives and annual sanitary controls. The purpose of these conditions is improving the living conditions and advancing in the social inclusion of the most vulnerable groups.
Likewise, the National Social Security Administration -ANSES- is responsible for the administration of PROGRESAR and the payment of its benefits. This institution also establishes the way to measure the family income of eligible young people. According to ANSES, a family group consists of the father and the mother if the young person is single or is composed of his partner if he is married or in a partnership. In all cases, the young's personal and family income must be less than or equal to the threshold quoted by the ANSES at the time of the request.5 Since its implementation in 2014 and until April 2015, this threshold was equal to the current minimum wage -MW-. Decree No. 505 of April 2015 raised the maximum level of the personal and family income to access the program to three times the current MW.
In addition to PROGRESAR, one of the most important labor market policies targeting young people is the Jóvenes con Más y Mejor Trabajo program. (Bertranou, Jiménez & Jiménez, 2018). This was created in 2008 to address the problems of unemployed young people who have not finished secondary school. Therefore, PROGRESAR allowed the extension of the benefits of this program to young people in employment and to those that completed high school (Bertranou & Casanova, 2015). However, there is a significant overlap between the eligible populations of both programs. PROGRESAR also partially shares the eligible population and the benefits of other current national programs of reinsertion in school and of college scholarships targeted young people (Bertranou, Jiménez & Jiménez, 2018).
The implementation of the PROGRESAR implied a significant increase in the potential coverage of active labor market policies for young people and a change in youth employment policy administration. The coverage of the program increased during 2014, surpassing the 500 thousand beneficiaries in September 2014 and reached 959 thousand young people in October 2015. Then, in 2016, the coverage dropped to almost 810 thousand participants and to about 782 thousand, in 2017 -Graph 1-. The number of beneficiaries made it in a program of massive scope among young people. In this way, PROGRESAR functions as an extension of the Asignación Universal por Hijo -AUH-6 program targeting children under 18. However, the coverage rate estimated by Bertranou, Jiménez & Jiménez (2018) decreased from 26.6% in the fourth quarter of 2014 to 18.3% of the total eligible youth population in the second quarter of 2016. This drop was due to the eligible population growth, because of the increase in the maximum level of income required to access in April 2015 and, to a lesser extent, to a decline in the program beneficiaries number.
Source: Authors' compilation based on ANSES (2015).
Graph 1 Evolution of the number of young beneficiaries of PROGRESAR, 2014-2016
One of the reasons for the relatively limited coverage of PROGRESAR might be the amount of the program's basic grant, which was originally 600 pesos and updated to 900 pesos in March 2015. Despite this increase, in 2016 PROGRESAR awarded grants equivalent to only 14% of the current minimum wage, 14.6% of the average labor income for young people and 19% of the average wage that a young person might earn from informal work (Bertranou, Jiménez & Jiménez, 2018).
Data and definitions
The microdata from the Permanent Household Survey -EPH- for the 2013-2015 period is the principal information source for the empirical analysis developed here. The EPH is an urban survey carried out by the National Institute of Statistics and Censuses -INDEC- of Argentina jointly with some provincial statistical offices since 1974. The general objective of the EPH is to know through indicators, a set of socio-economic dimensions of the population such as demographic, occupational, educational, housing and migratory characteristics. Thus, the EPH includes a specific labor market questionnaire following international standards, to guarantee certain comparability. In addition to the activity, employment, and unemployment rates, the EPH allows knowing some characteristics of the job position associated with its quality (Bertranou, Casanova & Sarabia, 2013). Therefore, it is the principal data source to characterize labor market functions.
The survey is conducted in the major urban centers of the country. The sampling in the urban areas covered by the EPH represents 71% of the urban population of Argentina and 62% of the total population.
Although the EPH is not a longitudinal survey, it is possible to construct this data type due to its rotation scheme known as 2-2-2. That is: two quarters into the sample, two quarters out and again two quarters into the sample, so each rotating panel is included to be interviewed in four times. In this way, a panel was built that allows monitoring of the individuals interviewed before -during the fourth quarter of 2013 and the first quarter of 2014- and after program implementation -during the fourth quarter of 2014 and the first quarter of 2015-.
Based on the information available in the EPH, salaried workers are classified into three job categories according to their quality. For this, we consider the possible combinations between the -in-formality condition and the deficits in other attributes such as work satisfaction, stability in employment and the extension of the working hours -over-occupation or involuntary part-time work-. Therefore, a worker has deficits in other job quality attributes if they are in one of the following situations: they have an unstable job -which includes a period of termination-, they are involuntarily employed part-time -they work less than 35 hours a week and want to work more-, they are over-occupied -they work more than 48 hours a week7- or declare dissatisfaction with their job -they are looking for another occupation or more hours of work-. In this way, salaried workers are classified as:
Formal salaried worker with deficits in other job quality attributes, that is, registered employees in at least one of the previous situations.
Informal salaried worker with deficits in other job quality attributes, that is, unregistered employees in at least one of the previous situations.
Formal salaried workers without deficits in other job quality attributes, that is, registered employees who are not in any of the previously listed situations.
In this context, the informal salaried workers with deficits in other job quality attributes have the worst employment in quality terms. On the contrary, the formal salaried work without deficits in other job quality attributes is considered the best quality occupation.
The EPH does not allow a similar analysis of the independent employment quality, mainly because it does not possess information on the social security coverage that these workers have. In any case, we consider this is not a significant limitation in this research because almost all the young people work in salaried positions -88%, on average- so that the empirical analysis leaves out a minor proportion of them -12%, on average-.
It is not possible to detect the PROGRESAR beneficiaries from the EPH data, but we can identify under certain premises the eligible population. Among the eligible population, we include all individuals satisfying the program's requirements, irrespective they actually participated in it -intention to treat-. Specifically, the eligible group is formed by young people between 18 and 24 with incomplete higher education in one of the following situations:
They are single living with their parents, and their total personal income or their parents' income does not exceed the program threshold.
They are single not living with their parents and their total personal income does not exceed the program threshold.
They are married or in a couple, and their joint income does not exceed the program threshold.
In addition to the unemployed and salaried young people -formal or informal-, those inactive are also eligible for the program. This is because, although Decree No. 84/2014 does not expressly include them, in practice it is very difficult to verify the condition of unemployment or inactivity.
On the other hand, the program design presents some problems to achieve a correct beneficiaries' selection according to its access requirements. For example, it is not clear how ANSES corroborates the level of income of informal employed young people or that of the relatives living with them -parents and/or spouses-. This highlights the program's difficulties to avoid the exclusion error as well as the inclusion error. Faced with these problems, we identify young people eligible for the program under the assumption that the ANSES can only corroborate the amount of income coming from a "registered" source -labor income of formal employment, retirement pensions, unemployment insurance or national subsidies-. Therefore, in this case, some young people with a personal income or with a family income higher than the threshold set in the program will be eligible, provided that the sum of their "registered" income does not exceed that limit. If the ANSES cannot corroborate all the access requirements, particularly those related to the amount of individual and family income of the applicants, the error of inclusion of the program could increase. However, it is likely that some of these young people decide to exclude themselves from PROGRESAR.
Table 1 presents some characteristics of the youth eligible for PROGRESAR by gender, obtained from the panel sample 2013-2015 corresponding to the period before and after its implementation. Regarding the activity status of the young people eligible for the program, it can be seen that, on average, most of them are inactive, while 21% are informal salaried workers and 10% are unemployed. However, in the case of young men, the majority of them is participating in the labor market. One issue to highlight is the increase in the proportion of eligible young employed as formal salaried after the program implementation.
Table 1 Characteristics of eligible young people for PROGRESAR, 2013-2015
Source: Authors' compilation based on INDEC, EPH.
On the other hand, there is a slightly larger proportion of men than women. Likewise, before program implementation, the majority of young people eligible had not completed secondary and belong to the two poorest quintiles of the family income per capita distribution. This indicates that PROGRESAR is focused on the most vulnerable young people. In addition, almost 26% of eligible youths declare that they do not study or work and most of them, in turn, do not look for work. While before the program, 26% dedicate their time only to work, after its implementation 32% are working.
Regarding the geographical distribution by residence region of the eligible, on average, 47% reside in the Greater Buenos Aires area -GBA-, 22% in the Pampa region and 14% in the Northwest of Argentina -NOA-.
On average, during 2013-2015, the majority of eligible youth are single -88%- and live with their parents -78% are head of household’s sons-. It should also be noted that more than 53% of these young people reside in households with children under 14 years of age that could be in their care. Therefore, we consider appropriate that the program offers assistance to beneficiaries from the Ministry of Social Development to find a child care center for their child and thus facilitate their reintegration into the education system.
According to the results of Table 1, there are some differences in the activity status of young people eligible for PROGRESAR before and after its implementation. However, there could be various factors affecting that change in young people’s employment situation. Therefore, to avoid the potential biases that could arise due to observable and unobservable differences between the treatment and control group across time, a fixed effect model is used.
Methodology
To analyze the impact of PROGRESAR on the transition probabilities of potential beneficiaries from unemployment to employment as well as towards quality jobs, the following linear probability model is estimated:
Where Y it is the outcome of interest, in this case, the probability that a young person finds a job or transits from a job with a labor quality deficit to another without any deficit, the variable T it is a dummy variable that identifies the individuals of the treatment group and takes the value 1 after the introduction of the program and the value 0, before that. The control group has a value 0 in this variable after and before the introduction of the program and is composed of all the young people who are not eligible but have the required age to access the program, who did not complete higher education and do not belong to the highest quintile. Then, it is the coefficient of interest that indicates the magnitude and the sign of PROGRESAR’s impact on the entry rate to a job and, in particular, to quality jobs. If labor transitions to quality jobs are affected by the program, the sign of the estimated coefficient for this variable will be significant. The vector of variables X it includes a set of individual controls that capture demographic and socioeconomic characteristics of the workers for adjust for observable differences between individuals that could bias the estimates of the interest coefficient. These controls include age -and its square-, educational level, marital status, position in the household, status of migrant, a dummy variable that indicate unsatisfied basic needs -NBI-,8 activity sector and size of the firm.9 It also includes dummies by region of residence.
The term α i is an unobserved random variable that is potentially correlated with the observed regressors X it . Thus the equation [1] is the fixed effect -FE- model. The individual-specific effect α i is unknown but is nonetheless possible to consistently estimate β in the FE model with short panels even though the conditional mean is not identified. In short panels the FE model permits identification of the marginal effect only for time-varying regressors. Thus the identification assumption in this case is thah α i is time-invariant and ε it is uncorrelated with X it and α i is correlated with X it (Cameron & Trivedi, 2005). In the case of PROGRESAR evaluation, it is being assumed that an individual’s propensity to participate in PROGRESAR may be endogenous, but the unobserved component of the effect of this propensity on labor transitions is constant over time once we control for observables X it .
As in the EPH it is not possible to identify PROGRESAR participants but only eligible youth: the coefficient β measures the intention to treat, that is, the impact that the intention to participate in a program has. If, in practice, there is a high level of completion of the treatment, it can be inferred that the ITT effectively measures the impact.
Likewise, the correct estimation of the causal effect of PROGRESAR requires the non-existence of another event different from the implementation of this program that generates a differential impact between the treatment group on the result of interest. In fact, this assumption seems to be fulfilled in the case of this article because in the first quarter of 2014, which separates the first observation period of the second period, no social, labor or economic policy measures were implemented, beyond the program with potential differential results between employees belonging to the treatment group and those classified in the control group.
As the dependent variable in [1] is binary, the estimated model is a linear probability model. Therefore, the analysis of the coefficients presents the typical limitations of this type of models. On the one hand, it does not guarantee that the probability predicted by the model is limited to the interval -0.1-. On the other, it assumes that the marginal effect of each explanatory variable is constant throughout its domain. However, the model is useful to determine the direction of the effects and reasonably approximates their magnitude for values of explanatory variables close to their sample means (Mario, Rosa & García, 2013). In addition, although non-linear models solve the mentioned limitations, they have many disadvantages, mainly when fixed effects are used. In particular, individual-specific effect α i is an incidental parameter if the panel is short as then each α i depends on fixed T observations and there are infinitely many α i since N →∞. This generates the incidental parameters problem that contaminates the estimation of the common parameters of interest -here β -. In general, the common parameters are also inconsistently estimated. Even where methods exist to consistently estimate β these methods tend to be model specific and no unified solution to the incidental parameters problem exists (Cameron & Trivedi, 2005).
Results of the impact evaluation
This section analyzes the impact of PROGRESAR on the young people’s transition rates from inactivity or unemployment to occupation and jobs of different quality. To this end, we estimate different specifications of the econometric model [1] described in the previous section. We analyze not only the impact for total eligible young but those for women and men were also estimated to find out if differences exist according to gender. The result of interest in this case is the coefficient of T it that measures the PROGRESAR marginal effect on dependent variables.10
In general, there are no significant effects of the program among eligible young unemployed -Table 2-. Instead the results indicate that the program affects the transitions that occur from inactivity to employment or between different quality jobs. The lack of a significant influence on the transitions from unemployment could respond more to the relatively low percentage of eligible youth who are initially in this state -10% according to Table 1- than to the absence of a program’s real impact on the exit rate of unemployment among young beneficiaries.
Table 2 PROGRESAR effects on eligible young people's access to different quality employment from unemployment
Source: Authors' compilation based on INDEC, EPH.
Note: Model 1 is the baseline model without control variables; the Model 2 includes individual characteristics such as gender, age, age squared, educational level, marital status and position in the household and Model 3 includes the variables mentioned in Model 2 and labor characteristics such as activity sector, qualification of the task, size of the firm and seniority in the occupation.
Table 3 PROGRESAR effects on eligible young people's access to different quality employment from inactivity
Source: Authors' compilation based on INDEC, EPH.
Note: Regressions control as described in the notes to Table 2.
The estimations suggest that PROGRESAR increases the probability of moving from inactivity to formal employment. In the full specification model, the positive impact on the probability of young people transiting from inactivity to a formal salaried employment amounts to 4.1 percentage points -p.p.-. In addition, this effect is greater among males -7.9 p.p.-, but it is not statistically significant among females. When considering the transitions to formal employment without deficit, we observe a positive estimated, statistically significant impact of 4 p.p. in the most complete model. The probability of moving from inactivity to formal jobs without deficits amounts to 4.9 p.p. among men and 5.3 p.p. among female. Nevertheless, these impacts are significant at the 0.1 level. Therefore, the results indicate that the program helps young eligible to access a formal employment but there is no significant impact on transition to informal jobs.
Regarding the entry rate to jobs of different quality, estimations indicate that PROGRESAR increases the probability of young people entering a formal salaried work. For the full sample, this positive impact varies from 3.4 p.p. in the most complete model to 4.2 p.p. in the baseline model. When comparing the estimates from men and women, we observe an interesting result. In this case, the program impact on entry to a formal salaried employment without deficit is only statistically significant among women -6.2 p.p.- but it is not between men. However, there is no program impact on the probability of entry into an informal salaried job.
Table 4 PROGRESAR effects on eligible young people's access to different quality employment from any employment status
Source: Authors' compilation based on INDEC, EPH.
Note: Regressions control as described in the notes to Table 2.
Conclusions
This article evaluated the impact of PROGRESAR on the labor transitions of young people aged 18 to 24 through a fixed effect model. Using the information available in the EPH database, we built a panel corresponding to the previous and later period to the program implementation.
The results obtained indicate that PROGRESAR produces an increase in the likelihood that eligible youth will transit to quality jobs. However, it may also be seen that there is no significant impact on transitions to informal employment. Also for young women initially not employed, the program increases their probability of entry to better jobs in terms of quality, such as a formal salaried employment without other deficits.
These positive impacts of PROGRESAR seem to be associated with the training and work experience activities included in the program. Several authors specifically argue that the dropout rate and low educational levels among young people are the main barriers preventing their labor insertion and trajectory towards a decent job (Bertranou, Jiménez & Jiménez, 2018). These findings show the need to implement or expand, as part of social protection systems, strategies that allow young people to reconcile paid work or study with domestic and care responsibilities.
The increase in young people's probability of getting a job without the quality deficit that the program produces could, in turn, generate indirect positive impacts such as the redistribution of income in favor of the poorest and the increase in economic growth from the rise in the productivity of the youth workforce. These positive results might be extended if the design and implementation of PROGRESAR is improved. Thus, for example, the low ratio between the beneficiary population and the eligible population highlights the need to increase the coverage rate and, maybe, the overlaps with other current school reinsertion programs and scholarships to pursue higher education for young people. Also, the lack of updating of the amount of the Programme's basic grant, in a context of still high inflation, could be one of the reasons for the low participation rate of young people in these programs (Bertranou, Jiménez & Jiménez, 2018).
Likewise, it is essential that young people with full-time employment do not face difficulties in attending to training centers. In this sense, it might include remote and online courses. A similar problem arises among the beneficiaries who have dependent children and need to combine their care tasks with the study. This is a relevant attribute considering that maternity and paternity constitute one of the main reasons for the lack of participation in the educational system by the young people that do not have paid work (Bertranou, Jiménez & Jiménez, 2018). In this context, it would be convenient for the Child Development Centers authorized by the Ministry of Social Development to carry out care tasks for the children of beneficiaries to be spread territorially throughout the country and between the different areas related to training. Another point to improve the design of the program is the inclusion of some requirement for passing courses or completing the current level of education, particularly if it is secondary education. The latter could not only increase skilled labor but also improve the living conditions of beneficiaries and their families.
Finally, the results should be interpreted with caution because the data used in this research present some important limitations, such as the difficulty in identifying the real beneficiaries of the program and following their labor trajectories over a longer period. This highlights the need to produce adequate databases such as administrative records of the programs with information of the beneficiaries before and after their implementation.
