Borrell Porta 2018a - Revision history

Scipediacontent at 13:56, 11 July 2018

2018-07-11T13:56:59Z

Scipediacontent at 13:52, 11 July 2018

2018-07-11T13:52:21Z

Danielquer: Danielquer moved page Review Borrell Porta 2018b to Borrell Porta 2018a

2018-04-08T08:13:28Z

Danielquer moved page Review Borrell Porta 2018b to Borrell Porta 2018a

Scipedia at 16:03, 5 March 2018

2018-03-05T16:03:49Z

Scipedia at 16:01, 5 March 2018

2018-03-05T16:01:40Z

Mireia: Mireia moved page Review Borrell Porta 2018a to Review Borrell Porta 2018b

2018-02-28T13:08:32Z

Mireia moved page Review Borrell Porta 2018a to Review Borrell Porta 2018b

Mireia at 13:06, 28 February 2018

2018-02-28T13:06:32Z

Show changes

Mireia: Mireia moved page Draft Borrell Porta 158029004 to Review Borrell Porta 2018a

2018-02-28T12:59:29Z

Mireia moved page Draft Borrell Porta 158029004 to Review Borrell Porta 2018a

Mireia: Created page with "

← Older revision		Revision as of 13:56, 11 July 2018
Line 15:		Line 15:

	''Note 1'': The source of all data, figures and tables that refer to the Panell de Desigualtats dataset is: PaD (Panel de Desigualtats Socials a Catalunya)- Fundació Jaume Bofill (2002 a 2012). They have been provided by the Fundació Jaume Bofill (FJB).		''Note 1'': The source of all data, figures and tables that refer to the Panell de Desigualtats dataset is: PaD (Panel de Desigualtats Socials a Catalunya)- Fundació Jaume Bofill (2002 a 2012). They have been provided by the Fundació Jaume Bofill (FJB).
−
−	~~''Note 2'': I certify that this is an original work from the author, owned by the author and that it has not been published elsewhere, nor will it be published before the Congress in May 2018.~~

	''Acknowledgements'': I thank Laura Morato from the Fundacio Jaume Bofill for all her support in providing the PaD dataset and answering all my queries and doubts, and Juli Carrere Balcells for his support in clarifying dataset-related questions on education.		''Acknowledgements'': I thank Laura Morato from the Fundacio Jaume Bofill for all her support in providing the PaD dataset and answering all my queries and doubts, and Juli Carrere Balcells for his support in clarifying dataset-related questions on education.

← Older revision		Revision as of 13:52, 11 July 2018
Line 13:		Line 13:
	This paper examines the impact of the Catalan 2004 law aimed at increasing public childcare coverage for under three years-old on the participation in formal childcare for families from different socio-economic backgrounds. To this purpose, it uses the ''Panell de Desigualtats Socials a Catalunya'' (PaD) database and exploits the uneven increase in public coverage across counties (comarques). The paper finds that while the policy alleviates the socio-economic participation gap between lower- and higher-income families, this one persists for certain sub-groups. More specifically, children whose parents have low education levels, or those from low-income families where the mother works part-time or is in charge of the house and children are significantly less likely to participate in formal childcare than their high-income counterparts.		This paper examines the impact of the Catalan 2004 law aimed at increasing public childcare coverage for under three years-old on the participation in formal childcare for families from different socio-economic backgrounds. To this purpose, it uses the ''Panell de Desigualtats Socials a Catalunya'' (PaD) database and exploits the uneven increase in public coverage across counties (comarques). The paper finds that while the policy alleviates the socio-economic participation gap between lower- and higher-income families, this one persists for certain sub-groups. More specifically, children whose parents have low education levels, or those from low-income families where the mother works part-time or is in charge of the house and children are significantly less likely to participate in formal childcare than their high-income counterparts.

−	~~Keywords:'' social investment, redistribution, Matthew effect, early childhood education and care, public policy.''~~

	''Note 1'': The source of all data, figures and tables that refer to the Panell de Desigualtats dataset is: PaD (Panel de Desigualtats Socials a Catalunya)- Fundació Jaume Bofill (2002 a 2012). They have been provided by the Fundació Jaume Bofill (FJB).		''Note 1'': The source of all data, figures and tables that refer to the Panell de Desigualtats dataset is: PaD (Panel de Desigualtats Socials a Catalunya)- Fundació Jaume Bofill (2002 a 2012). They have been provided by the Fundació Jaume Bofill (FJB).

@@ Line 21: / Line 21: @@
 ''Acknowledgements'': I thank Laura Morato from the Fundacio Jaume Bofill for all her support in providing the PaD dataset and answering all my queries and doubts, and Juli Carrere Balcells for his support in clarifying dataset-related questions on education.
-. Introduction
+== 1. Introduction ==
 Socio-economic gaps in cognitive and non-cognitive behaviour start at an early age, even before school begins (Francesconi and Heckman 2016). Already at the age of three, children from professional families have a wider range of vocabulary than families in welfare or working-class families (Heckman and Mosso 2014). These early life skill gaps can be explained by the importance of early years to the development of certain parts of the brain, such as speech production or higher cognitive functions (Shonkoff and Phillips 2000). At the same time, research also shows that disadvantages experienced at an early stage are difficult to be fully overcome at later stages in life (Heckman and Mosso 2014).
@@ Line 37: / Line 37: @@
 The next section reviews the 2004 Catalan policy and at the same time it puts it within the broader context of childcare provision both in Spain and the European Union. Section 3 describes the data and methods, section 4 presents the results, followed by section 5 that shows some robustness checks. Finally, section 6 discusses the findings and concludes.
-:<big>'''2. Early childhood education and care in Catalonia'''</big>
+== 2. Early childhood education and care in Catalonia ==
 The 2004 law approved in Catalonia (law 5/2004, 9 July), stipulated that at the end of the 2004-2008 period, at least 30,000 new public childcare settings for under-threes had to be created and maintained thereafter, prioritizing disadvantaged areas, respecting the territorial equilibrium and taking demand into account (Sindic de Greuges 2007). This law was the response to a popular legislative initiative from 2002 that demanded an increase in the number of public childcare settings to fulfil the increasing demand (ibid.).
@@ Line 50: / Line 50: @@
 ''[Insert Figure 1 here]''</div>
-:<big>'''3. Data description and empirical strategy'''</big>
+== 3. Data description and empirical strategy ==
 <big>'''3.1 Data'''</big>
@@ Line 93: / Line 93: @@
 The strategy adopted is one that tries to control for other potential sources that can be confounded with the increase in public childcare coverage. First, the analysis includes county dummies, so that the coefficients are estimated within each county. In this way, we make sure that the effect that we see is ''not'' the effect of different public childcare coverage levels in different counties, but rather, the ''increase'' in public childcare coverage ''within'' a certain county. Second, we include a time trend. This is because it is plausible that during the decade for which we carry out the analysis, the participation to formal childcare changes, say, because of higher female labour force participation rates, or because of changing norms, or due to any other variable that may follow a linear trend across time. The chosen strategy does not deal with problems of reverse causality – participation in formal childcare driving increase in public coverage - and does not fully deal with the problem of omitted variable bias. Having said that, given that it is the first study of this kind, I believe the added value is still very significant.
-:<big>'''4. Results'''</big>
+== 4. Results ==
 Table 2 below presents the main results. All regressions show the average marginal effects of our main independent variables – an increase in public childcare coverage and family income - on the probability of attending formal childcare arrangements. Model 1 shows the effect of the main independent variables, with no interactions and no controls. Model 2 interacts the two main variables, model 3 adds socio-economic controls and model 4 adds demographic and institutional controls. All models include a time trend and county dummies.
@@ Line 152: / Line 152: @@
 ''[Insert Figure 6 here]''</div>
-:<big>'''5. Robustness checks'''</big>
+== 5. Robustness checks ==
 In this section, some robustness checks with reference to the dependent, independent variables, sample and methods are carried out. Table 6 summarises the results.
@@ Line 172: / Line 172: @@
 To increase the security that attrition does not lead to problematic estimates, in model 8 the sample is restricted to the cross-sectional observations and the first observations of the panel individuals. This restricted sample, however, has a lower average child age than the full sample, with the former being, on average, less than one, compared to 1.25 in the full sample. With regard to income levels, the restricted sample has 24% of families in the lowest income, whereas the full sample has 21%, which is not a significant difference. The same happens with education levels, with the restricted sample having 12% of parents with lowest level of education, compared to 10% in the full sample. Indeed, it seems that the largest difference between the two samples is the age of the child, and results in Panel A confirm it, displaying certain similarities in the coefficients with those in model 2, where only children aged 0 and 1 where included. The estimates in Panel B, albeit different from those in the full sample, are not strikingly different, with the confidence intervals overlapping. From these results we can conclude that attrition is not likely to significantly alter our results.
-:<big>'''6. Discussion and conclusion'''</big>
+== 6. Discussion and conclusion ==
 This paper has examined the impact of increasing public childcare coverage in Catalonia during the 2000s. The findings suggest that, while the policy was a much needed step towards the right direction - alleviating the socio-economic participation gap between lower-income and higher-income families - this one persists for certain sub-groups. More specifically, the paper has found that even at higher public childcare coverage levels, children whose parents have lower education levels are significantly less likely to participate in formal childcare than their more educated counterparts, regardless of their income. Similarly, low-income families where the mother works part-time or is in charge of the house and children also fall short of participating in formal childcare at equal rates than their higher-income counterparts.
@@ Line 180: / Line 180: @@
 At the same time, the paper maybe raises more questions than it answers. The finding that even with higher coverage levels the families with lower-educated parents still participate less in formal childcare than their more educated counterparts raises questions on ''what factors are exactly at play. ''As argued in the paper, education is linked to higher opportunity costs, more liberal values on gender roles, and higher levels of information and awareness of the benefits of early childcare education and care. The present paper cannot disentangle which of these factors (or other) are part of the underlying cause for the persistence of the Matthew effect. Unless we understand the root causes that lead lower-educated families to participate less in formal childcare, we will keep raising a society that is, in its foundation, unequal. Similarly, while the link between not working full-time and lower participation rates seems strong for lower-income classes, the paper doesn’t analyse the direction of the relationship. Is the decision to work part-time endogenous and related to the willingness to care for their children? If so, why don’t we see the same participation rates for mothers in charge of the house chores and part-time mothers from higher-income families? These and other related questions should be tackled in subsequent research.
-<big>'''Bibliography'''</big>
+== Bibliography ==
 ALMOND, D. and CURRIE, J. (2011) “Human capital development before age five”. In O. Ashenfelter and D. Card (Eds.), Handbook of Labor Economics, Volume 4B, Chapter 15, pp. 1315–1486. North Holland: Elsevier.
@@ Line 234: / Line 234: @@
 <big>'''<br/>'''</big>
-<big>'''Tables and Figures'''</big>
+== Tables and Figures ==
 '''Table 1 '''Descriptive statistics

@@ Line 9: / Line 9: @@
 -->
-<big>'''Abstract'''</big>
+== Abstract ==
 This paper examines the impact of the Catalan 2004 law aimed at increasing public childcare coverage for under three years-old on the participation in formal childcare for families from different socio-economic backgrounds. To this purpose, it uses the ''Panell de Desigualtats Socials a Catalunya'' (PaD) database and exploits the uneven increase in public coverage across counties (comarques). The paper finds that while the policy alleviates the socio-economic participation gap between lower- and higher-income families, this one persists for certain sub-groups. More specifically, children whose parents have low education levels, or those from low-income families where the mother works part-time or is in charge of the house and children are significantly less likely to participate in formal childcare than their high-income counterparts.
@@ Line 21: / Line 21: @@
 ''Acknowledgements'': I thank Laura Morato from the Fundacio Jaume Bofill for all her support in providing the PaD dataset and answering all my queries and doubts, and Juli Carrere Balcells for his support in clarifying dataset-related questions on education.
-:<big>'''1. Introduction'''</big>
+. Introduction
 Socio-economic gaps in cognitive and non-cognitive behaviour start at an early age, even before school begins (Francesconi and Heckman 2016). Already at the age of three, children from professional families have a wider range of vocabulary than families in welfare or working-class families (Heckman and Mosso 2014). These early life skill gaps can be explained by the importance of early years to the development of certain parts of the brain, such as speech production or higher cognitive functions (Shonkoff and Phillips 2000). At the same time, research also shows that disadvantages experienced at an early stage are difficult to be fully overcome at later stages in life (Heckman and Mosso 2014).

← Older revision	Revision as of 08:13, 8 April 2018
(No difference)

← Older revision	Revision as of 13:08, 28 February 2018
(No difference)

← Older revision	Revision as of 12:59, 28 February 2018
(No difference)