Assessing the Impact of a Minimum Income Scheme in the Basque Country

Transcription

1 Assessing the Impact of a Minimum Income Scheme in the Basque Country Sara de La Rica University of the Basque Country and FEDEA Lucía Gorjón University of the Basque Country and FEDEA (Preliminary Draft: Do not Cite) Abstract: In this paper, we evaluate the impact of a Minimum Income Scheme (MIS), which operates in one of the 17 Spanish Regions - the Basque Country for more than two decades. In particular, we test whether the policy delays entry into the labour market for scheme recipients. In addition, we test the efficacy of policies directed to activate beneficiaries of the MIS so as to re-entry into the labour market. Our results indicate that the Minimum Income Scheme, on average, in addition to preventing social exclusion through financial support, does not delay labour market entry. However, the impact is heterogeneous across demographic groups. Furthermore, Active Labour Market Policies designed for this group, and in particular training, do have a strong positive impact for finding a new job. JEL Classification: C14, C21, C52. Key words: minimum income schemes, active labour market policies, poverty, inverse probability weighting, propensity score matching

2 1. Introduction As of today, most European Member states provide some form of Minimum Income Scheme so as to ensure a minimum standard of living for households when they lack other sources of financial support. The emergence of these schemes dates back to 1992, when the European Council recommendation assessed the need of developing last resort schemes, which recognised the basic right of every individual to ensure a decent minimum standard of living. These programs were part of comprehensive and consistent plans to combat social exclusion 1. Implementation of Minimum Income Schemes (MIS) across European Countries has been since then heterogenous in coverage and effectiveness. The widest ones are the so-called "Simple and comprehensive schemes", which basically cover every person/household in need of support, without confining to particular categories of people. Since 2008, The European Council endorsed the objective of combining adequate income support with labour market activation measures so as to facilitate relocation of beneficiaries into the labour market. Although the implementation of these schemes is progressing in most European countries, albeit heterogeneously, there is no sufficient assessment of their impact on aspects such as poverty reduction, labour market participation of beneficiaries, and/or the impact of activation measures of their beneficiaries on their relocation. Examples of these studies include Gouveia and Rodrigues (1999) and Brunori, Chiuri and Peragine (2009), who evaluate the impact of particular Minimum Income Schemes on Poverty Reduction, implemented in Portugal and in a southern Italian region, respectively. Additionally, Clavet, Duclos and Lacroix (2013) and Chemin and Wasmer (2012) assess the impact of two Minimum Income Schemes - one potentially implemented in Quebec and the other in Alsace-Moselle, in France, on the Labour Market Participation of their beneficiaries. Surprisingly, we are not aware of any study that assesses the impact of activating labour market measures for MIS beneficiaries on their labour market relocation. Our paper tries to fill this gap. Specifically, our study evaluates the impact of a Minimum Income Scheme that operates in a northern Region of Spain - The Basque Country. This region was a pioneer at the introduction of a Minimum Income Scheme in Spain in As of today, the Basque Country is the only Spanish region with a 1 For more details, The Council Recommendation 92/441/EEC of 24 June 1992: 6bf88a5407f3/language-en

3 Simple and Comprehensive Scheme of MIS 2. We assess, on the first hand, the extent to which the Basque Minimum Income Scheme delays entry into the labour market for their beneficiaries. Second, we test the efficacy of policies directed to activate their beneficiaries for labour market relocation. We do so by using the Inverse Probability Weighting methodology, which allows beneficiaries of the MIS to be compared to a similar fictitious collective created by the weighting of non-beneficiaries. By doing so the treatment is dissociated from the individual characteristics and hence, the treatment is pseudo-randomized. Our results indicate that, on average, the Basque Minimum Income Scheme does not, per se, delay labour market entry for their beneficiaries. Interestingly, however, the impact differs across demographic groups. Furthermore, Active Labour Market Policies designed for this group, and in particular training, do have a strong positive impact for their labour market relocation. The remaining of this paper is organised as follows: in section 2 we revise institutional aspects of the Minimum Income Scheme implemented in the Basque Country. Section 3 briefly describes related literature. In section 4, data description and main descriptives of Minimum Income Scheme recipients are presented. Section 5 is devoted to methodological and analytical evaluation methods and to present the empirical findings. Finally, section 6 summarizes and concludes. 2. The Minimum Income Scheme in the Basque Country Region The Basque Minimum Income Scheme was introduced in 1989, with the so-called Integral Plan to Combat Poverty 3. Over the last decades it has gone through several modifications. In 1998 it becomes a law, concepts of poverty and exclusion were defined and employment incentives, penalties or infractions were established. Moreover, the quantity and the requisites to be beneficiary have also been modified several times. The last modification was implemented in 2011 (Law 4/2011). We base description details in that version. Eligibility Requisites: A first important aspect to note is that the Basque Minimum Income Scheme is household based. To apply for the aid, applicants must comply with the following eligibility requisites: First, they must show that their household income is insufficient to meet basic needs, which means impossibility of access to the goods and services of a minimum welfare in the society according to the Basque Government 2 The Basque Country is a small region in the northeast of Spain. The population reaches 2 million inhabitants (the 5% of the Spanish population). Active Labour Force exceeds 1 million people and occupation rate is 50%. The Basque Country is among the richest Spanish Regions, ranking second in GDP per capita and third in low unemployment rate (12,8%). In addition, the Basque Human Development Index is 0,924, the highest in the country, and at the level of the Netherlands. 3 Different laws can be found here:

4 criterion of poverty, which is described below. A second eligibility condition concerns residency in the Basque Country: In principle, the holder of the MIS in the household must be registered in the census and have actual residence in the Basque Country over the last three years without interruption. If they can prove five years of paid work experience in the Basque Country, the residence requisite can be relaxed to one year of residence instead of three. If none of the above requirements are met, they must have been registered for five continuous years from the immediately preceding ten years 4. Furthermore, the Minimum Income Scheme is understood as a last resort scheme, and as such, it is required that all other Income Aids to which they are entitled to have been previously requested. In principle, the scheme is compatible with other income aids or salaries of any family member, meanwhile it does not exceed the defined poverty line. In addition, it is required that there is no more property other than the habitual residence. Coverage: MIS is transferred monthly to family units on a monthly basis. The amount setting by the Basque Government that addresses the basic needs varies depending on the minimum wage (MW), the number of people in the household, the number of retired and whether it is a single-parent family or not. In particular, it is set as 88% of the MW for single member household and it reaches until 125% of the MW for a household with three or more members. In the case of households with at least one pensioner the percentages rise to 100% and 135% respectively. Single parent families receive a complementary subsidy 5. If there are other incomes in the household, the MIS covers the difference of that amount. Household Labour Market Availability: Both holders and other beneficiaries cohabiting in the same family unit who are able to work are committed to being available to do so. In addition, they have to participate in activities that increase their employability. In particular, the holder must sign an inclusion-oriented employment improvement agreement. However, although the spirit of the law is that every beneficiary should have an active job search, only around 40% are observed to receive any public employment service or activating intervention 6. We are not aware of the criteria that Public Employment Service uses to follow MIS perceivers in order to control for their activation, i.e., we ignore whether the individuals are self-selected into different activities or there are some kind of compulsory participation. 4 For particular groups, such as those who receive a public pension or have been victims of domestic abuse, there is no need to prove work experience and there is only the condition of one year of residence in the Basque Country. 5 Concretely, the amount in 2015 varies from euros, for a single member household, until euros, for a household with three or more members and at least one pensioner. Single-parent families receive 45 extra euros. 6 In particular, the activation rate for holders is 41,7% and for non-holders 38,2%.

5 3. Related Literature Very similar policies have been implemented, although few actually evaluated, in other countries. Furthermore, some pilot projects can also be found and ex-ante or ex-post evaluations of similar measures to reduce poverty throughout the world. The Portuguese Guaranteed Minimum Income (Rendimento Minimo Garantido) set up in 1996 follows a very similar structure to the Basque Renta de Garantía de Ingresos. However, the reference income for the benefit is very low, approximately 50% of the absolute poverty line. Gouveia and Rodrigues (1999) provide a simulation of its effect on poverty using the Household Budget Survey 1994/95. They find a participation of 5% of the households and 5.7% of the individuals in the programme. The public cost is about 0.18% of the Portuguese GDP and 0.39% of total public expenditure, but the impact on the beneficiaries income increase on average by 18.5% in the annual income participated households. According to this study, this policy measure has a modest effect on the reduction of the number of poor households, but has a relevant effect on the intensity and severity of poverty. Following Beckerman s model that analyses the efficiency of income transfers, Rodrigues (2001) estimates the vertical efficiency of the programme (as the proportion of total transfers received by the households that were poor before the transfers) and the poverty reduction efficiency of the programme (as the proportion of total transfers that contributed to a reduction in poverty). This study concludes that the vertical efficiency of the programme is 85% and the poverty reduction efficiency is 82%. This means that households that were initially above the poverty line receive 15% of the total MIS transfers, and 18% of total transfers did not contribute to a reduction in the poverty gap. A similar analysis has not yet been done for the Basque Country, but it is actually in progress. Reddite Minimo d Insermento pilot experiment was implemented in a small town in the south of Italy, Mola di Bari. Brunori, Chiuri and Peragine (2009) analyse such issues as (i) eligibility criteria, (ii) targeting choices and results, (iii) distribution and welfare effect on the beneficiaries and on the overall population in the town and (iv) the incentive effects on the labour market participation. The most relevant result they find is that the simple use of a national measure of poverty tends to obscure individual situations, making it impossible to distinguish poor people in need of public support. The second relevant issue opposes this, suggesting that local administrations seem unable to correctly verify the income level of households. This study also shows which categories of individuals are more subject to be activated. The analysis of the MIS shows how a relevant number of households improve their economic conditions thanks to the public transfer, even though there are insufficient coverage rates. Finally they did not find evidence to suggest that individuals involved in the programme tend to also become beneficiaries in the following months.

6 Ex-ante evaluation of a Proposal in Qébec is developed by Clavet, Duclos and Lacroix (2013). Every individual would be guaranteed an income equivalent to 80% of the Market Basket Measure. He first estimates a structural labour supply model and next he simulates the impact of the reduction- poverty recommendation by the Qebec Commité. By predicting labour supply the result shows that the proposed scheme would have strong negative impacts on the labour market participation rates and mostly among low-income workers. In France we find the so-called Reveneu Minimum d Insertion evaluated by Chemin and Wasmer (2012) in Alsace-Moselle in the eastern part of France. Their estimates, based on double and triple differences, show that the RMI policy was associated with: a 3% fall in employment (among unskilled workers between 25-55), leading to an estimated loss of 328,000 jobs; a decline in the jobaccess rate; and a five month increase in the average duration of unemployment. They also find considerably larger disincentive effects for single parents. 4. The Dataset and Some Descriptive Statistics 4.1. The Dataset Our dataset consists of monthly longitudinal information of all individuals who are registered in the Basque Public Unemployed Service from February 2015 to January The data is collected the last day of each month. Most registered individuals are unemployed, but some may be employed while searching for another job. Their labour market status is clearly stated. All Minimum Income Scheme beneficiaries and their cohabitants must be registered in the Public Unemployment Service as a requisite to receive the income aid, independently of their labour market status. The database includes all information provided by every individual when registering at the Unemployment Office, such as the standard demographic characteristics (gender, age, educational level, nationality, postcode and residence, knowledge of other languages), as well as labour market information (previous labour market experience, occupational and geographical search, unemployment duration, etc.). The Basque Public Unemployment Service also provides exact information on whether individuals receive or have received unemployment benefits (either entitled benefits, assistance benefits and/or MIS) as well as the length of entitlement. Finally, the database also registers information about the public employment services unemployed workers have received in the last 12 years to enhance job access. Information such as the type of service, number of hours and the date of start and end are provided. Public Employment Services in Spain divide the pool of unemployed workers included in their files into "Registered Unemployed" and "Other Unemployed Workers". The latter category, which accounts for around 22% of all unemployed workers, includes retired or pensioners, those not immediately available for a job, those registered in the

7 current month, those who just seek particular kinds of work, such as outwork, teleworking, or those who seek jobs under 20 hours a week. Students are also included in this category. We restrict our analysis to the group of Registered Unemployed Workers, those without a job, seeking for it and immediately available for any "regular" job. With respect to Minimum Income Scheme beneficiaries, the Basque Country has around MIS beneficiaries each month in the period under analysis, which accounts for 25% of the registered-unemployed people in the Basque Country Statistical Distribution of MIS Recipients vs non-mis Recipients To get a precise idea of the compositional differences between MIS recipients and the rest of registered unemployed workers, we present the distribution of each of the two groups under the following four characteristics altogether: gender, age (<30, and >44), educational level (primary education at most, secondary, and university level) and duration of unemployment (<3 months, 3-6 months, 6-12 months, and >24 months). Figures 1 and 2 display the distribution of the MIS and non-mis recipients, respectively, along the four characteristics. We do so for a particular month - October 2015, to get a better idea, not only on relative but also on absolute terms. Any other month of the sample would present almost identical patterns. At first sight, it can be seen that the profile according to educational level and unemployment duration is quite different among recipients of MIS and the rest of the unemployed. This is not surprising given that this income scheme is understood as a last resort scheme. In particular, 60% of MIS recipients do not have secondary studies and more than a half have spent more than two years in an unemployment situation. These figures barely reaches one third and one fourth for the non-mis recipients, respectively. To be more precise, the most prevalent group among recipients is the one with very long-term unemployment, over 30 years and primary education level. They represent a third of the total MIS recipients. Comparing with the group of Non- MIS recipients, the collective with these characteristics barely reaches 10%. Furthermore, regardless of educational level, MIS individuals over 30 who have spent more than two years looking for a job account for 50%. If we now focus on the youngest group, we see that more than half have spent more than two years seeking employment and the vast majority (70%) have only primary education. However, the pattern is very different among those who do not perceive the MIS, where those unemployed for a very short duration are generally young people with secondary or higher education. It is important to note that many young people with higher education levels continue studying if they do not find a job and, therefore they are not considered unemployed. This behaviour is not found between unemployed people with lower educational levels, which are precisely the more common in the MIS group.

8 4.3. Monthly Exit Rates from Unemployment to Employment (Job Finding Rates) We now proceed to describe the patterns of monthly job finding rates for beneficiaries and non-beneficiaries of the MIS, exploiting the longitudinal nature of our dataset. We define exit into the labour market as a transition from registered-unemployed in the previous month to an employed labour status in the current month. Following the same structure as before, we characterise job-finding rates comparing beneficiaries with non-beneficiaries of the MIS by the same four characteristics than before, i.e, gender, age, educational level and unemployment duration. Given that we observe unemployed people from February 2015 to December 2015 we compute job-finding rates from March 2015 to January On average, the monthly job finding rate for MIS beneficiaries is 3%, significantly lower than the one observed for non-mis recipients, which amounts to 9%. Figures 3 and 4 display job finding rates for MIS and non MIS recipients, respectively, along different profiles. It is immediate to observe that for the two groups job finding rates increase with educational level and strongly decrease with unemployment duration. To give some numbers on the strong negative association between unemployment duration and job finding rate, Figures 3 and 4 show that those unemployed workers whose unemployment duration is lower than three months face an average exit rate of 11%, while for the group of the very long-term unemployed (over two years) on average it only reaches 1%. Interestingly, 60% of MIS recipients exhibit these very long-term unemployment durations. Another point to note is that, although the level of education is relevant to understanding differences in access to a job, it is far removed from the importance of the unemployment duration: the exit rate of the MIS collective with higher education reaches on average 5%, compared with 2% of those with primary education. Figure 4 focuses on the comparison between job finding rates between MIS recipients and non-mis recipients. As mentioned before, on average there is a difference of 6 percentage points between job finding rates of the two groups. However, this difference varies remarkably depending on the personal profiles. For example, among the very short duration unemployed there is a difference of 7.5 points, while among the very long duration unemployed the difference barely reaches one percentage point. [Insert Figure 3 and 4 here]

9 4.4. Determinants of the probability of finding a job: MIS recipients vs non-recipients Finally, we estimate the probability of finding a job the last day of each month for all those unemployed people registered on the last day of the previous month. As before, we calculate the probability of finding a job from March 2015 to January The dependent variable, therefore, takes value 1 if the unemployed person gets a job in the next month, and 0 otherwise. To perform this exercise we take into account all observable variables that may affect the employability of people registered in the Public Employment Service. In particular, we include demographic characteristics such as sex, age, nationality, disability, education and language skills; job characteristics such as requested occupations, experience, activity in the previous field of work, unemployment duration, geographical scope of the new job search, month(s) in which the individual is observed as unemployed, whether the individual is the MIS holder or not 7, and province of registration. We add a dummy indicating whether the individual has ever been derived to social services. The collection of benefits in the current or in the previous months is also included. We include in our estimation an indicator for whether the individual has gone through any activating service at least once in the last six months. 40,7% of MIS recipients have received any kind of service in the last six months, as compared to 13,75% for the MIS non-recipients 8. We divide the type of activating service into: guidance, monitoring, information of self-employment and training. Table 1 presents the results of the estimation - marginal effects are displayed - using a pooled probit with month and province fixed effects 9. The first column estimates the probability of finding a job for the MIS group whereas the second one presents estimates for MIS non-recipients. Note this estimation does not account for unobserved heterogeneity. It should be taken as a preliminary view of the importance of the characteristics of the unemployed people in the job search process. The most remarkable result has been already anticipated: Unemployment duration is the variable that most affects the probability of leaving unemployment. The chances of entering into the labour market are reduced dramatically as the time the person remains unemployed increases. The largest decrease in the probability of getting a job occurs after the barrier of 3 months (reference group) with a reduction of 5 percentage points when the individual stays unemployed between 3 and 6 months. 7 Note that the MIS is provided at households level, whereas the development of the analysis is done at individual level. The data does not permit identify which individuals belong to the same household, only whether they are MIS holders or not. 8 These figures correspond to October. Any other month of the sample present similar percentages. 9 Population average with a robust estimator of the variance is used.

10 Furthermore, being unemployed between 6 months and 1 year reduces the likelihood by one point (up 6.5 points compared to being less than 3 months unemployed) and between 1 and 2 years duration the probability falls 1.6 points (up 8 points). The negative impact increases to 9 points if the duration of unemployment goes beyond 4 years. If we compare these results with the impact of the same variable on the total unemployed people non-percipient of MIS (column 2), we see that the duration of unemployment has also the greatest negative impact. In particular, spending more than 3 months unemployed reduces the exit probability by almost 8 percentage points. As occurs with the MIS collective, the exit likelihood continues declining with the duration of unemployment, reaching 15.7 points if 4 years of unemployment are exceeded. As it can be seen, there is no other variable that has a similar impact. Considering the level of studies, we see that in general the likelihood of finding a job correlates with the educational level of the unemployed individual: Having secondary education (compared with primary or no education) increases the probability midpoint; having completed high school increases by 0.8 points; medium vocational training rises by 1.2 points and higher vocational training and high university degree by 1.9. Notice that the impact of being more that 3 months unemployed is double than having university studies (as compared to primary or no education) for MIS beneficiaries. Given that a separate section is dedicated exclusively to a counterfactual evaluation of the impact of Activating Services on the probability of finding a job, here we present only a preliminary assessment of the activating interventions. It is important to note that the information for self-employment has a clearly differentiated nature, since people who use it are practically in motion towards self-employment. Thus, measuring the effectiveness of it by the impact on the probability of leaving for a job does not make much sense. From now on, we evaluate the effectiveness of only the other three interventions in the exit into employment. Training is by far the most effective service for both collectives (MIS receivers and non-receivers). Individual guidance increases their employability as well. However, monitoring does only increase the probability of finding a job for MIS collective and not for the rest of the unemployed workers. [Insert Table 1 here] 5. Assessing the Impact of the Basque Minimum Income Scheme on the Labour Market: A Counterfactual Evaluation Any Minimum Income Scheme is, in itself, a Passive Policy, as its main aim is to guarantee all individuals the necessary resources for satisfying their minimum needs. However, as mentioned before, the Basque Minimum Income Scheme, following the dictates of the European Council since 2008, requires (in principle) recipients to

11 participate in active policies to make their reintegration into the labour market as fast and successful as possible. Given the double-scope of the MIS in its passive and active vein, our evaluation of the policy is also twofold. On the first hand, although the goal of any passive policy is not to accelerate the employability of the unemployed, but supplement their income to alleviate their poverty, empirical evidence generally finds that most income transfers to the unemployed produce a delay in job finding rates. Reservation wages increase for anyone who receives additional income, and this typically delays job entry, hence reducing job-finding rates. However, there are two aspects of the MIS which might accelerate rather than delay job access: First, given that the MIS can also be enjoyed by employed workers with insufficient income for satisfying minimum needs, MIS beneficiaries might be willing to accept jobs with "low" remunerations - compatible with retaining the transfer. Second, the MIS can be removed if it is proved that the beneficiaries reject job offers. For these reasons, the typical "delay" effect of a passive transfer like the MIS may be partially offset by some kind of acceleration effect for reasons other than the activating measures implemented on them. Our first assessment with respect to the impact of the MIS in the Basque Country is whether MIS causes either delay or acceleration effect and if so, measure its magnitude. This will be the first objective to be addressed in this section. On the second hand, and possibly more interesting, is to assess whether active policies offered to the MIS beneficiaries enable a better transitions towards employment. This will be the second objective of the section Empirical Evaluation Strategy In the two analyses, the aim is to evaluate the impact either of the MSI itself or rather of activating measures directed to the MSI collective, on the probability of leaving unemployment. As in previous estimations, the dependent variable (Y ) takes the value 1 if the unemployed individual gets a job in the next month and 0 otherwise. The treatment (D ), which is a dummy variable, takes value 1 firstly when the individual is MIS beneficiary and secondly if the individual receives active policies 10. The covariates included in our analyses are the same than in previous estimations (X). The main problem we face in both analyses developed in this paper is sample selection. In the first one, unemployed people need to comply with strict requirements to become MIS beneficiary. In the second analysis, the profile of the unemployed people that receive activation services differs broadly with respect to non-receivers characteristics (as it will be shown later). Consequently, given that individuals are not 10 We are not able to develop a duration analysis because of data limitations. Our dataset includes only 2015 information and, for MIS beneficiaries, where more than a 70% have been unemployed for more than one year, we would need longer longitudinal information.

12 randomly chosen, the use of a mean difference between the outcomes of treated and control group to know the causality of the corresponding treatment is not valid. In the case where treatment participation depends on observable characteristics (X) only, we can estimate the Average Treatment Effect on the Treated (ATT) by conditioning on these variables, rendering the counterfactual outcome is independent of the treatment (conditional independence assumption, CIA). However, the probability of finding a job for the MIS and MIS non-receivers might be affected by confounding factors. Therefore, the validity of CIA in this analysis is difficult to justify. In the second analysis, given our misunderstanding of the selection process of being service participant, we are not able to argue whether CIA is satisfied of not. Propensity Score methods are useful to estimate treatment effects using observational data since they allow observational studies to be designed similar to randomized experiments (Rubin, 2001). Rosenbaum and Rubin (1983) show that instead of conditioning on the covariates, conditioning on the probability of potential treatment conditional on observable covariates, the propensity score (p x = P (D = 1/X)), suffices to achieve balance between treatment and control group as long as other requirements are satisfied. Firstly, the covariates influencing assignment and outcome do not have to predict deterministically the treatment participation (weak overlap, P D = 1/X < 1for all X). Secondly, treatment participation of one individual cannot have an impact on the outcome of other treated or control individual. Our two samples validate the weak overlap. Furthermore, it seems reasonable to think that being MIS beneficiary or service perceiver does not affect the probability of finding a job of other individual. For these reasons, we find appropriate the using of Propensity Score techniques. Different Propensity Score approaches have been suggested to estimate and adequate counterfactual outcome, where the predominant used methods are matching or reweighting (Imbens, 2004). These methods try to remove observed systematic differences between treated and control subjects. In our first analysis, the Inverse Probability Weighting (IPW) makes the distribution of observable covariates similar between the treated and control group. 11 Furthermore, as it will be justified below, IPW is the only valid methodology in our first analysis due to the characteristics of the treatment. For the second research, given the lack of knowledge of the selection mechanism and the characteristics of the evaluated sample, we calculate the treatment effect following two different methodologies: the Inverse Probability Weighting (IPW) and the Propensity Score Matching (PSM). The idea behind the Inverse Probability Weighting is the following: Random assignment guarantees that the distribution of the covariates among units of observation in the treatment and control groups is probabilistically equivalent, i.e., all 11 Table 2 is presented later on to show the distribution of the characteristics in the reweighted sample.

13 units are equally likely to be in the treatment or control groups. When the assignment is not random, however, some individuals are more likely to be treated than others, depending on their particular characteristics. To account for these differences in the regression formulation we should weigh observations according to the inverse probability of receiving treatment. Thereby, a pseudo-random sample is obtained by weighting observations by the inverse of the probability of being treated. Therefore, the distribution of covariates among the groups would be probabilistically equivalent (see Gardeazabal and Vega-Bayo, 2015). Summarizing, weighting individuals by the inverse probability of treatment received creates a synthetic sample where treatment assignment is independent of the observed covariates. Inverse Probability Weighting allows obtaining unbiased estimates of average treatment effects. However, these estimates are only valid if there is no residual systematic differences in observed variables between the weighted treated and control groups (Austin and Stuart, 2015), as we prove to be the case. In this manner, it is assumed that reducing the observable differences, the unobservable factors are also diminished. It stands to reason that a more efficient estimator could be obtained if the regression of the reweighted sample includes all measured covariates as additional regressors. This other estimator is known as Augmented Inverse Probability Weighting (AIPW). The IPW estimator uses a two-step approach to estimate the treatment effects. The specification for the Average Treatment Effect on the Treated (ATT) is as follows: 1) Estimate the probability of being treated based on the covariates by a probit 12 regression. It is denoted p! (x), i.e., the propensity score. Use the inverse probability weights to compute the new pseudo-random sample. Build regression weights (w! ) as: w! = 1 if D! = 1 w! = p!(x) 1 p! (x) if D! = 0 The idea behind this reweighting procedure is quite straightforward. Its objective is to approximate the distribution of the covariates of the control group to the treated group. For that reason all treated individuals keep their weights equal to 1. Control individuals with a probability of being MIS beneficiaries that equal 0.5 would weigh 1; those with a probability higher than 0.5 would weigh more than 1 with an increasing pattern and those with a probability lower than 0.5 would weigh less than 1 with a decreasing pattern. In doing so, the outcome of those control individuals with higher probability of being MIS beneficiaries would weigh gradually more and the outcome of those control individuals with lower probability of being MIS beneficiaries would weigh exponentially less. 12 Logit model can be also used.

14 2) Calculate the ATT of the new sample. That is to say, run a probit regression of the outcome on a constant and the treatment using the calculated weights. The coefficient of the binary treatment in the previous regression is a consistent estimation of ATT, provided that the propensity-score is correctly specified. Adding all measured confounders as additional covariates the Augmented Inverse Probability Weighting (AIPW) estimator is obtained. In the second evaluation an additional Propensity Score approach is applied. On this occasion, the Propensity Score Matching (PSM) helps us to estimate the impact of the activation services as well. This methodology entails matched sets of treated and untreated subjects who share a similar value of the propensity score (Rosenbaum and Rubin, 1985) and it allows estimating the ATT (Imbens, 2004). The most common implementation is one-to-one pair matching, in which pairs of treated and control individuals are formed in such a way that they have similar values of propensity score. Once a matched sample has been formed, the treatment effect can be estimated by directly comparing outcomes between treated and control matched individuals. Schafer and Kang (2008) suggest that, within matched sample, the treated and control subjects should be regarded as independent. In contrast to this, Austin (2011) argues that the propensity score matched sample does not consist of independent observations. He maintains that in the presence of confounding factors, covariates are related to outcomes and thus, matched subjects are more likely to have similar outcomes than the ones presented by randomly selected subjects. Based on Austin argument we discard the use of the Propensity Score Matching in the first analysis. Non-observed factors, such as family income, differ systematically between the treated and control individuals as it is a crucial determinant to be selected for the treatment. However, we will develop the second evaluation using PSM as we find reasonable to argue that the unobservable factors of treated and control individuals resemble more each other (given the selected control group used) than in the first analysis Impact of MIS on job finding rates - Does MIS delay the probability of job finding? As we have seen in previous sections, Minimum Income Scheme beneficiaries face a monthly job finding rate of 3%, while for the non-mis unemployed group it reaches 9%. However, as already stated, the composition of the MIS beneficiaries differs notably from the rest of the unemployed, and these compositional differences (longer unemployment duration and lower educational level, primarily) can cause at least part of these observed differences in job finding rates. In order to isolate compositional differences from the income scheme, we use the Inverse Probability Weighting Methodology, detailed above, so that we can assess the extent to which the observed difference in job finding rates is explained by (i) compositional differences between the two groups and (ii) by the income scheme.

15 To do so, we include in the treatment group all those individuals who are beneficiaries of the MIS in the current month. Given that the observation unit is one individual per month, it can occur that one person belongs to the treatment group in some months, when they benefit from the MIS, but do not belong to it others - those where he/she is not receiving it. Hence, it can be the case that one individual belongs to the treated group in a concrete month and to the control group in a different month. To create an adequate counterfactual, we must define the control group so that it simulates as best as possible job finding rates from the group of MIS beneficiaries had not they received such a benefit. According to the data, 93% of MIS beneficiaries receive ONLY this type of income aid; another 6% also receive other type of welfare benefit and the remaining 1% receives contributory benefits. In the last two situations, they receive both types of income aids because the received benefits are still lower than what it is considered necessary to meet basic needs in the household. We think it makes sense to deduce that if the income scheme did not exist, this 93% of actual income beneficiaries would not be getting any additional income aid and the remaining 7% would receive an insufficient amount. For this reason, we have chosen to include in the control group those unemployed individuals who do not receive ANY benefit in the current month 13. For this group, the observed monthly job finding rate is 6.5%. Consequently, the outcome of the evaluation must be interpreted as the differential impact of MIS in the job finding rate with respect to not receiving any benefit. However, the treatment (receiving MIS) is by no means random. We have previously specified that there are specific requirements, some of them observable in our dataset but others non observed confounder variables, such as total income in the household, that we need to control for. To "correct" for these differences between the treatment and the control group we use the Inverse Probability Weighting method. Table 2 present the distribution of the reweighted control group that validates the use of the IPW methodology. This table shows that the differences in the main characteristics have been eliminated using the explained weighting procedure. [Insert Table 2 here] The results of the Inverse Probability Weighting Estimation, as well as an extended version of it, the Augmented Inverse Probability Weighting Estimator, are presented in Table 3. Applying the mentioned methodology, the result is that the impact of MIS is not significantly different from zero for any significance level. The result is the same for both the IPW and the AIPW estimators, which make it more reliable 14. This indicates that the monthly job finding probability for Minimum Income Scheme beneficiaries 13 In October this group consists of a total of 69,961 unemployed, compared with unemployed beneficiaries of MIS. 14 The evaluation has also been developed using the Propensity Score Matching methodology. However, the results are divergent, corroborating the Austin (2011) argument.

16 would have been the same if they had not received any benefit. So we can conclude that the Minimum Income Scheme itself does not delay the job finding probability. In other words, the observed differences in job finding rates between the treatment and the control group are solely due to the difference in the composition of both collectives and not caused by the effect of the policy. [Insert Table 3 here] As a second step, we analyse whether the MIS has different impact for different demographic groups. Specifically, we evaluate the MIS impact for males and females separately, three groups of age (< 30, and > 45) and three groups of education (primary, secondary and tertiary) 15. Results are presented in Table 4 and confirm that the impact of MIS is not homogeneous across demographic groups. In particular, for women MIS delays exit to employment slightly (0.2 p.p) whereas the impact is null for males. Second, the MIS accelerates job finding for older workers (0.2 p.p) whereas for young workers (<30) the impact is just the opposite as it delays exit to employment (1 p.p). Finally, we find a delay impact of MIS for low educated workers (0.2 p.p), whereas the MIS accelerates job entry for those with more than primary education (0.2 p.p if workers with secondary education and 0.5 p.p for tertiary educated workers). [Insert Table 4 here] Our results coincide partially with the ex-ante evaluation in Clavet, Duclos and Lacroix (2013) and with the findings (double and triple differences estimation strategy) in Chemin and Wasmer (2012). Both find a negative impact on labour market participation, concretely among specific collectives, such as low-skill workers. However, their results are not directly comparable to ours as the methodology and the design of the policies in the regions examined by these studies. To our knowledge there is no comparable evaluation of a similar policy. The main conclusion of this exercise is as follows: The MIS by definition reduces poverty and promotes social cohesion. Our analysis leads to the conclusion that on average the MIS itself does not delay exit for a job. However, we do find differences in the impact for different demographic groups. In particular, it causes an undesired delay effect commonly found in other passive policies on women, low educated and young people, while it accelerates the entry into the labour market for medium and high-educated workers as well as for those over 45 years. 15 Same analysis is not developed by duration of unemployment because of the endogeneity of the variable.

17 5.3. The Impact of Active Policies on job finding probability for Minimum Income Scheme beneficiaries In this section we evaluate the effectiveness of the activating interventions received by the MIS beneficiaries. Such evaluation is highly recommended given that in general, active policies are quite costly. Evaluating them allows to test and if necessary modify to improve the efficiency of the Public Employment Service in providing their recipients with the necessary tools to relocate into the labour market. This information can certainly highlight what actions should be strengthened, modified or even eliminated. As a reminder, we are focusing in three types of active policies: guidance, monitoring and training. First of all, we denote that a person uses any of the three types of services if, at least once in the last six months, including the current one, it is observed that the service was received. Secondly, we present some descriptive statistics to show the extent of activation for the MIS group. As it is done in the descriptive section - in order to present the characteristics of the unemployed - we focus on a particular month (October 2015) so as to avoid overrepresentation of the long-term unemployed. Of the unemployed people registered as MIS beneficiaries in that month, people have received some kind of active policy, either orientation, monitoring or training at any time in the last 6 months. This amounts to 40,8% of the total. Focusing on the type of service they have received, people (39,4% of all unemployed MIS recipients) have received guidance services, 265 (0.7%) monitoring services and 881 (2.3%) training courses. In other words, there are some people who have received more than one type of service (exactly 728 people). Given the low frequency of monitoring, from now on, we focus the results on activation through guidance or training interventions. Below we present a brief profile of how individuals involved in each of these two types of policies are compared to individuals who do not receive any activation intervention. Table 5 displays the distribution of the four main characteristics (sex, age, education and unemployment duration) depending on the type of active policy they receive. [Insert Table 5 here] In general men participate in these interventions more than women: around 65% of participants in training are males. The age range varies depending on the type of service they engage in. Orientation and training is higher mainly for those in the age range - their relative incidence among MIS receivers is 46%. In general, young people tend to receive fewer activation interventions. Considering the educational level there are also substantial differences: 60% of MIS beneficiaries have at most primary education, 27% secondary and 13% a higher education, which means that on

18 average, activation is lower for highly educated MIS recipients. In addition, activation measures decrease with unemployment duration. Furthermore, we find distributional differences by activation type. Although orientation measures distribute similarly by educational level, we find significant differences in the training programmes, as those with secondary or a higher educational level use more training services than those with at most primary education. To evaluate the impact of each of these activation interventions, we classify into the treatment group those MIS beneficiaries that have received the particular activation policy under evaluation - either individual guidance or training - in the last six months. As before, we measure the impact of receiving such activation service on monthly job finding rates. We use as control group the MIS beneficiaries who have not participated in ANY activating service from the Public Unemployment Services in the last six months in order to get a cleaner impact of each specific activation service. As such, the results must be interpreted as the impact of the intervention on job finding probability as compared to not receiving any activating service in the last six months. As depicted in Table 5, the treated and control group differ in the composition of important characteristics such as the duration of unemployment or educational level. Therefore, we evaluate each intervention following the IPW methodology described previously. In that way, the interventions are "pseudo-randomised", so that the distribution of the covariates between the two groups is balanced and the treatment would be probabilistically equivalent. Therefore, the impact of each type of intervention can be properly assessed without the results being biased to differences in composition. In addition to the IPW (and AIPW) methodology, we also use a Propensity Score Matching Technique to enhance robustness. Given that now the control group consists of MIS-beneficiaries (although they do not receive activation measures), we find reasonable to assume that unobserved confounding factors of treated and control individual do not broadly differ between the two groups. This assumption is essential to validate the use of the Propensity Score Matching technique. The results of the evaluation of each of the two active policies for MIS beneficiaries - guidance and training, are shown in Table 6. Inverse Probability Weighting (IPW), Augmented Inverse Probability Weighting (AIPW) and the Propensity Score Matching (PSM) estimators are presented. The first three columns correspond, respectively, to each of the three specifications for the impact of guidance service. It can be seen that guidance exerts a positive impact on exit into employment. This impact is statistically significant for the three methodological approaches, although its magnitude differs slightly across them. As a general result, we can conclude that guidance increases the

19 probability of getting a job in about half percentage point compared to those that have not received any type of activating intervention in the last six months 16. The last three columns in Table 6 present the impact of training programmes on job finding rates. Unfortunately, we do not have information neither on the type of training nor on whether there is any previous selection process to participate on a training program. Hence, all we can assert given this information limitation is whether participating in any training program helps finding a job. And the result we find is that training is undoubtedly the one with highest impact in the probability of finding a job for the MIS group. Individuals who use these programmes increase their job finding likelihood by around 3 percentage points. Given that the average job finding rate for MIS beneficiaries is 3%, the job finding probability increases by around 100% when a MIS unemployed attends a training course. Due to its potential to find a job, it would be very convenient to have more detailed information with regards to training programs so as to evaluate in the future more precisely which types of training programs seem to work better. [Insert Table 6 here] According to the literature in Active Labour Market Policies, our results also find that an adequate design in the activation policies accelerates the return into the labour market. Summarising, active policies accelerate significantly the job finding probability for the Minimum Income Scheme beneficiaries. However, only around 40% of these individuals use them even though in principle it is stated that participating in them is required. Specifically, training is the most effective policy, doubling the chances of finding a job for those who undergo it. This conclusion emphasises the importance of linking passive policies with active policies, given that, in comparison with the unemployed people that do not receive any aid, those MIS beneficiaries that use active policies enhance their chances of finding a job. 6. Summary and Conclusions In the Basque Country (a north-east region of Spain) a Minimum Income Scheme has been continuously implemented since Its main objective is to guarantee all individuals the resources for covering their basic needs, while at the same time providing for their progressive social and professional integration. Furthermore, in line with the European Council recommendations, the Basque MIS has an interesting feature, i.e., the recipients are in principle required to participate in active policies to make their reintegration into the labour market as fast and successful as possible. 16 The impact of the guidance has also been addressed for the population subgroups. The results are not shown here given the lack of interest as all profiles present similar results. Training programmes are not evaluated for the different population collectives due to sample size reasons.

20 In 2015, there were about 62,000 beneficiaries of the Minimum Income Scheme, of which 60% belonged to the group denoted by registered-unemployed by the Public Employment Service. The rest are workers, retired beneficiaries, or non-working persons that for different reasons do not fit into the category of registered unemployed citizens. MIS beneficiaries account for 25% of all the registered unemployed in the Basque Country. Given that the Basque Minimum Income Scheme is a last resort one, low educational levels and (very) long-term unemployed are widespread among beneficiaries. Specifically, 60% of MIS beneficiaries have, at most, primary education and 52% spent more than two years looking for a job. Unsurprisingly, low educational levels and particularly long unemployment durations are the determinants that strongly delay job finding. Indeed, MIS beneficiaries face an average 3% monthly job finding rate, while for the other unemployed but non-receivers of such aid, this rate stands at 9%. The first empirical strategy of this paper is to measure whether this difference is solely due to the different composition of the unemployed or, on the contrary, if the MIS delays the entry into the labour market, as empirical evidence has proven that in general passive policies do. The second aim of the paper is to measure the effectiveness of the active policies on the MIS beneficiaries in terms of the impact in the job finding probability. Even though all MIS recipients should engage in activation activities, the fact is that only around 40% individuals (16,000 out of 38,000 unemployed beneficiaries) have done so at any time in the last six months. Guidance is the most common service: 39% of all the MIS unemployed beneficiaries receive it, followed distantly by the training services - only 2.3% participate in them. The profiles of the participants differ between the kind of activation they receive and also with respect to those who do not participate in these services. To develop both evaluation exercises, propensity score methods are applied. In both analyses we follow an Inverse Probability Weighting methodology. Furthermore, in the second exercise, we complement our evaluation with a Propensity Score Matching. Both methodologies help us contend with the confounding effects and the different composition between the treated and the control groups in the most suitable way according to the characteristics of the corresponding sample. Our results confirm that on average the Minimum Income Scheme does not delay entry into the labour market and hence the difference in the observed job finding rates are only due to the different composition of the treated and control group. If the analysis is developed for specific population groups, we find that its impact differs. The undesired delay effect commonly found in passive policies is observed in low educated and young MIS beneficiaries. On the contrary, the MIP accelerates the entry into the

21 labour market for medium and high-educated people and for unemployed over 45. To the best of our knowledge there are no comparable evaluations of a similar policy implementation that we could compare our results with. The second finding indicates that all types of public services have a positive impact on job finding rates. However, not all services affect the exit probability equally: the most effective services are the training programmes, which double the probability of finding a new job, followed by guidance services, which increase the probability by around 20%. Hence, as a policy device, this study drives the conclusion that training services for MIS beneficiaries should be enforced, as they help recipients to relocate into the labour market, which is the final aim of activating initiatives. Moreover, it is essential to emphasize the importance of linking passive policies with the activation of the beneficiaries. Finally, for future research, we have two related projects to work on, both of which require more information. The first one is to extend our study to a duration type analysis, where the question to answer is not based on instantaneous job finding rates, but rather, on time to exit from unemployment. By now we are limited by the fact that we have information of all unemployed workers but only for 12 months. For MIS beneficiaries, where more than a 70% have been unemployed for more than a year, we would need longer longitudinal information. Second, we would like to get more precise information on the training programs MIS workers receive, so as to learn more on which types of training programs are more successful for job finding rates, and so, be able to be more precise with regards to policy advice.

22 References Alvarez, J., & Arellano, M. (2003). The Time Series and Cross-Section Asymptotics of Dynamic Panel Data Estimators. Econometrica, 71(4), Angrist, J. D., & Pischke, J. S. (2008). Mostly harmless econometrics: An empiricist's companion. Princeton university press. Austin, P. C. (2011). An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate behavioral research, 46(3), Austin, P. C., & Stuart, E. A. (2015). Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Statistics in medicine, 34(28), Ballester, R. (2012). Los programas de garantía de rentas en España: la renta mínima de inserción catalana y sus componentes de inserción laboral (No ). Departamento de Economía-Pontificia Universidad Católica del Perú. Brunori, P., Chiuri, M. C., & Peragine, V. (2009). The Economic effects of a Local Minimum Income Support Program (No. 029). SERIES Working Paper. Caliendo, M., Künn, S., & Schmidl, R. (2011). Fighting youth unemployment: The effects of active labor market policies. Card, D., Kluve, J., & Weber, A. (2010). Active labour market policy evaluations: A meta-analysis. The Economic Journal, 120(548), F452-F477. Chemin, M., & Wasmer, E. (2011). Ex-ante and ex-post evaluation of the 1989 French welfare reform using a natural experiment: the 1908 social laws in Alsace-Moselle. under review at the Journal of Political Economy. Clavet, N. J., Duclos, J. Y., & Lacroix, G. (2013). Fighting Poverty: Assessing the Effect of Guaranteed Minimum Income Proposals in Québec.(No. 7283). IZA Discussion Paper. Colombino, U., Locatelli, M., Narazani, E., & O Donoghue, C. Alternative Basic Income Mechanisms: An Evaluation Exercise. Cordazzo, P. (2005). Minimum Income Allocation System (RMI) A Longitudinal View. Evaluation review, 29(5), de la Rica, S. (2015). Políticas activas de empleo: una panorámica. Fedea Policy Papers- 2015/01.

23 Ehlert, C. R., Kluve, J., & Schaffner, S. (2012). Temporary work as an active labor market policy: evaluating an innovative program for disadvantaged youths. Ferrera, M. (2005). Welfare states and social safety nets in Southern Europe.Welfare state reform in southern Europe. Fighting poverty and social exclusion in Italy, Spain, Portugal and Greece, Freire, G. B. O. (1997). La renta mínima garantizada: a caballo de la Asistencia, la Seguridad Social y la Política de Empleo. Revista de treball, economia i societat, (7), Gardeazabal, J., & Vega Bayo, A. (2016). The economic cost of armed conflict. Gouveia, M., & Rodrigues, C. F. (1999). The impact of a" Minimum Guaranteed Income Program" in Portugal. Heckman, J. J., LaLonde, R. J., & Smith, J. A. (1999). The economics and econometrics of active labor market programs. Handbook of labor economics,3, Imbens, G. W. (2004). Nonparametric estimation of average treatment effects under exogeneity: A review. The Review of Economics and Statistics, 86, Imbens, G. W., & Wooldridge, J. M. (2009). Recent developments in the econometrics of program evaluation. Journal of economic literature, 47(1), Kluve, J., Card, D., Fertig, M., Gora, M., Jacobi, L., Jensen, P.,... & Schmidt, C. M. (2007). Active Labor Market Policies in Europe. Kluve, J. (2010). The effectiveness of European active labor market programs. Labour economics, 17(6), Marchal, S., Marx, I., & Van Mechelen, N. (2016). Minimum income protection in the austerity tide. IZA Journal of European Labour Studies, 5(1), 1. Matsaganis, M., Ferrera, M., Capucha, L., & Moreno, L. (2003). Mending Nets in the South: Anti-poverty Policies in Greece, Italy, Portugal and Spain. Social Policy & Administration, 37(6), Ravagli, L. (2015). A minimum income in Italy (No. EM16/15). EUROMOD at the Institute for Social and Economic Research. Rodrigues, C. F. (2001). Anti-poverty effectiveness and efficiency of the Guaranteed Minimum Income Programme in Portugal. Rosenbaum PR, Rubin DB. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika; 70:41 55.

24 Rosenbaum PR, Rubin DB. (1984). Reducing bias in observational studies using subclassification on the propensity score. Journal of the American Statistical Association; 79: Rosenbaum, PR, Rubin, DB (1985). Constructing a control group using multivariate matched sampling methods that incorporate the propensity score. The American Statistician, 39, Rubin DB. (2001) Using propensity scores to help design observational studies: application to the tobacco litigation. Health Services & Outcomes Research Methodology; 2: Rubin DB. (2004) On principles for modeling propensity scores in medical research. Pharmacoepidemiology and Drug Safety; 13(12): Whitton, T. (1993). Does' insertion'work? France's minimum income. Citizen's Income, (19), pp-11. Wooldridge, J. M. (2015). Introductory econometrics: A modern approach. Nelson Education. Wooldridge, J. M. (2010). Econometric analysis of cross section and panel data. MIT press.

25 Figures Figure 1 Unemployed MIS beneficiaries in the Basque Country, October 2015 Men, <30 Men, Men, >45 0 1,000 2,000 3,000 4,000 5,000 < >24 < >24 < >24 Women, <30 Women, Women, >45 0 1,000 2,000 3,000 4,000 5,000 < >24 < >24 < >24 Source: Own elaboration. Lanbide microdata Primary Secondary Tertiary Figure 2 Unemployed Non-MIS beneficiaries in the Basque Country, October 2015 Men, <30 Men, Men, >45 0 5,000 10,000 15,000 20,000 < >24 < >24 < >24 Women, <30 Women, Women, >45 0 5,000 10,000 15,000 20,000 < >24 < >24 < >24 Source: Own elaboration. Lanbide microdata Primary Secondary Tertiary

26 Figure 3 Figure 4