Impact of MGNREGA on Private Coaching in West Bengal, India
Private coaching and the impact of
the rural employment guarantee
programme on it: Evidence from
West Bengal, India
Upasak Das
December 2016
Introduction
•
Sustainable Development Goals: Achieving universal access to
quality education.
•
Widespread literature on provision and access of education and
its effects
(Lóapez and Valdés, 2000; McCowan, 2007; Lewin
2009)
•
One dimension of education that has been of importance is the
choice and expenditure on private coaching
•
Private coaching: Fee-based tutoring that is given to provide
supplementary instruction to students in academic subjects that
they study in the mainstream academic system (Dang and Rogers,
2008)
•
Malaysia, Japan, South Korea: Majority of the students in primary
as well as middle and secondary level receive private coaching.
•
Private coaching is more used by higher income households,
residing in urban areas, studying in higher grades or appearing for
competitive examinations,
low achievers or studying in
government schools (
Baker et al. 2001, Glewee and Jayachandran,
2006, Glewwe & Kremer, 2006, Tansel and Bircan, 2006, Dang,
2007; Kim and Lee, 2010, Dongre and Tewary, 2014,
Gangopadhyay and Sarkar 2014).
Objectives
•
Impact of the Mahatma Gandhi National Rural Employment
Guarantee Act (MGNREGA) on educational expenditures on
private coaching in parts of West Bengal
•
If participation and extent of participation indicated by the
number of days of work under MGNREGA and the earnings from it
have an impact on the decision to send the children for private
coaching and the expenditure incurred on it
NOTE:
More than 53% of the sampled households who worked in
MGNREGA reported that the money received has been spent on
private coaching for their children apart from food and clothing
MGNREGA
•
The MGNREGA was passed by the Indian Parliament on 23
rd
August 2005
•
At least 100 days of guaranteed wage employment (unskilled
work) in every financial year to every rural household at a
statutory minimum wage.
•
Welfare impacts
–
Reducing poverty and enhancing welfare (
Imbert & Papp, 2014;
Klonner
and Oldiges 2014
)
–
Reduction in short-term migration (Das, 2015)
–
Empowerment of women (Khera and Nayak, 2009; Dev, 2011)
–
Access to credit (Dey and Imai, 2015)
–
Positive impact of female participation on educational outcomes (Afridi et
al. 2016)
Private Coaching in India and West Bengal
•
NSSO, 2014:
–
About 25% of the students depend on private coaching
–
At primary level, 23% of the male students and 20% female
students opt for private coaching
–
At higher levels, this increases to 38% and 35% for male and
female students respectively
•
Azam (2015): Private coaching is
relatively inelastic at each levels
of schooling and hence is considered as
a necessary good in the
household consumption basket
•
West Bengal: More than 65% opt for private coaching. At higher
levels, this is about 90%
Data
•
Field survey conducted from January to April 2012 in two blocks
of Cooch Behar district of West Bengal
•
Two blocks- Haldibari Block and Cooch Behar-I block
•
Haldibari block
: Dakhin Bara Haldibari GP and Devanganj GP;
Cooch Behar-I block
: Dawaguri GP and Falimari GP
•
556 households were surveyed
•
Questions of socio-economic and demographic characteristics;
MGNREGA participation, private coaching expenditure (monthly).
Outcome and Primary Variables of Interest
•
Children of 6 to 18 years.
•
Outcome Variable: whether the child attends private coaching and
expenditure incurred on it [Log (expenditure+1)]
•
Primary Variables:
–
Participation in MGNREGA in 2011 (any member got work)-
binary.
–
Log (number of days of work in MGNREGA in 2011 +1)
–
Log (annual earnings from it +1)
•
Controls: Age, gender, socio-economic, demographic factors.
Estimation Methodology
•
Regression equation:
if child, from household, in GP, is sent to
private coaching, 0 otherwise/ expenditure incurred.
and = child specific and household characteristics
= household participation in MGNREGA
•
Omitted variable bias (Angrist and Pischke, 2009)
•
OLS regression of participation/ extent of
participation in MGNREGA.
•
First stage regression:
=
vector of exogenous variables which affect
participation in MGNREGA
= the vector of instruments defined at GP level
•
Putting the residuals in the second stage regression
•
Attendance in private coaching: probit regression
Expenditure incurred: OLS and tobit regression
•
Bootstrapping at the village level to correct for the
standard errors
•
If the coefficients of the residuals are found to be
statistically significant at 1% or 5% level, the variable
should be treated as endogenous (Ravallion and Wodon,
2000)
Instruments
(i)
A dummy to indicate if the GP is ruled by the
Trinamool Congress (TMC) Government
(ii) Expenditure incurred for unskilled workers at the
GP level per 1000 households demanding work
under MGNREGA and
(iii) Number of MGNREGA staffs employed at the GP
level in 2015-16
First stage regression
Descriptive Statistics
•
Over 52% of the children attends private coaching
Regression Results (Participation)
Regression Results (Number of working days)
Regression Results (Earnings)
Local Polynomial Regression Plots
Qualitative Evidence
•
An
agricultural labourer from the Dawaguri
GP
•
Private coaching popular in Dawaguri GP-
about 70% of the children attends it
•
Households from GPs in the haldibari block
Conclusion
•
Evidence of impact of MGNREGA in increasing the
expenditure on education through private coaching
•
Large welfare impacts through MGNREGA
•
Glaring problems making MGNREGA less attractive
(
Mukhopadhyay et al., 2015, Narayanan et al. 2016)
•
Need to address problems related to rationing of jobs
and delayed payments
•
Higher allocation of resources to the programme
This study explores the impact of the Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) on private coaching expenditures in West Bengal, focusing on how participation and earnings under MGNREGA influence decisions related to private tutoring. More than half of households involved in MGNREGA reported using the earnings for private coaching, shedding light on the relationship between rural employment programs and educational expenditures.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
E N D
Presentation Transcript
Private coaching and the impact of the rural employment guarantee programme on it: Evidence from West Bengal, India Upasak Das December 2016
Introduction Sustainable Development Goals: Achieving universal access to quality education. Widespread literature on provision and access of education and its effects (L apez and Vald s, 2000; McCowan, 2007; Lewin 2009) One dimension of education that has been of importance is the choice and expenditure on private coaching
Private coaching: Fee-based tutoring that is given to provide supplementary instruction to students in academic subjects that they study in the mainstream academic system (Dang and Rogers, 2008) Malaysia, Japan, South Korea: Majority of the students in primary as well as middle and secondary level receive private coaching. Private coaching is more used by higher income households, residing in urban areas, studying in higher grades or appearing for competitive examinations, low achievers or studying in government schools (Baker et al. 2001, Glewee and Jayachandran, 2006, Glewwe & Kremer, 2006, Tansel and Bircan, 2006, Dang, 2007; Kim and Lee, 2010, Dongre and Tewary, 2014, Gangopadhyay and Sarkar 2014).
Objectives Impact of the Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) on educational expenditures on private coaching in parts of West Bengal If participation and extent of participation indicated by the number of days of work under MGNREGA and the earnings from it have an impact on the decision to send the children for private coaching and the expenditure incurred on it NOTE: More than 53% of the sampled households who worked in MGNREGA reported that the money received has been spent on private coaching for their children apart from food and clothing
MGNREGA The MGNREGA was passed by the Indian Parliament on 23rd August 2005 At least 100 days of guaranteed wage employment (unskilled work) in every financial year to every rural household at a statutory minimum wage. Welfare impacts Reducing poverty and enhancing welfare (Imbert & Papp, 2014; Klonner and Oldiges 2014) Reduction in short-term migration (Das, 2015) Empowerment of women (Khera and Nayak, 2009; Dev, 2011) Access to credit (Dey and Imai, 2015) Positive impact of female participation on educational outcomes (Afridi et al. 2016)
Private Coaching in India and West Bengal NSSO, 2014: About 25% of the students depend on private coaching At primary level, 23% of the male students and 20% female students opt for private coaching At higher levels, this increases to 38% and 35% for male and female students respectively Azam (2015): Private coaching is relatively inelastic at each levels of schooling and hence is considered as a necessary good in the household consumption basket West Bengal: More than 65% opt for private coaching. At higher levels, this is about 90%
Data Field survey conducted from January to April 2012 in two blocks of Cooch Behar district of West Bengal Two blocks- Haldibari Block and Cooch Behar-I block Haldibari block: Dakhin Bara Haldibari GP and Devanganj GP; Cooch Behar-I block: Dawaguri GP and Falimari GP 556 households were surveyed Questions of socio-economic and demographic characteristics; MGNREGA participation, private coaching expenditure (monthly).
Outcome and Primary Variables of Interest Children of 6 to 18 years. Outcome Variable: whether the child attends private coaching and expenditure incurred on it [Log (expenditure+1)] Primary Variables: Participation in MGNREGA in 2011 (any member got work)- binary. Log (number of days of work in MGNREGA in 2011 +1) Log (annual earnings from it +1) Controls: Age, gender, socio-economic, demographic factors.
Estimation Methodology Regression equation: = + + + + + + Y X H D u w ijk ijk jk jk i j k i j = k if child, from household, in GP, is sent to private coaching, 0 otherwise/ expenditure incurred. 1 Y ijk X H and = child specific and household characteristics jk ijk = household participation in MGNREGA D jk
Omitted variable bias (Angrist and Pischke, 2009) OLS regression of participation/ extent of participation in MGNREGA. First stage regression: = + + + D C Z jk jk k jk = vector of exogenous variables which affect participation in MGNREGA C jk kZ = the vector of instruments defined at GP level
Putting the residuals in the second stage regression Attendance in private coaching: probit regression Expenditure incurred: OLS and tobit regression Bootstrapping at the village level to correct for the standard errors If the coefficients of the residuals are found to be statistically significant at 1% or 5% level, the variable should be treated as endogenous (Ravallion and Wodon, 2000)
Instruments (i) A dummy to indicate if the GP is ruled by the Trinamool Congress (TMC) Government (ii) Expenditure incurred for unskilled workers at the GP level per 1000 households demanding work under MGNREGA and (iii) Number of MGNREGA staffs employed at the GP level in 2015-16
First stage regression Probit Tobit Working in MGNREGA Log of number of days worked Log of earnings Instrument Variables GP ruled by TMC -1.512*** -3.331*** -8.754*** (0.312) (0.610) (1.778) Number of MGNREGA staffs at GP level 0.490*** 1.267*** 2.990*** (0.110) (0.203) (0.576) Expenditure incurred per 1000 workers at GP level 0.173*** 0.498*** 1.115*** (0.049) (0.088) (0.252) Controls Yes Yes Yes N 387 387 387
Descriptive Statistics Over 52% of the children attends private coaching Variables Total No private Attend private Difference coaching coaching 194.251 Expenditure on private coaching (in Rs.) 0.494 0.489 0.499 -0.010 Participated in MGNREGA in 2011 (P) 15.882 12.593 18.813 - 6.220* Number of days worked in MGNREGA (2011) 2100.376 1621.163 2527.447 - 906.284* Annual income (2011) (in Rs.)
Regression Results (Participation) Going to tuition (Probit) Expenditure in tuition (OLS) Expenditure in tuition (Tobit) Without residual With residual Without residual With residual Without residual With residual Residuals -1.060* -2.475** -4.195** (0.579) (0.963) (1.655) Worked in MGNREGA 0.223* 1.283** 0.421** 2.896*** 0.760** 4.956*** (0.115) (0.570) (0.190) (0.945) (0.384) (1.657) Controls Yes Yes Yes Yes Yes Yes GP fixed effects Yes Yes Yes Yes Yes Yes N 739 739 739 739 739 739
Regression Results (Number of working days) Going to tuition (Probit) Expenditure in tuition (OLS) Expenditure in tuition (Tobit) Without residual With residual Without residual With residual Without residual With residual Residuals -0.172 -0.440* -0.723* (0.137) (0.229) (0.394) Log of working days 0.109** 0.281** 0.194*** 0.635*** 0.362*** 1.085*** (0.045) (0.131) (0.075) (0.221) (0.135) (0.370) Controls Yes Yes Yes Yes Yes Yes GP fixed effects Yes Yes Yes Yes Yes Yes N 739 739 739 739 739 739
Regression Results (Earnings) Going to tuition (Probit) Expenditure in tuition (OLS) Expenditure in tuition (Tobit) Without residual With residual Without residual With residual Without residual With residual Residuals -0.105 -0.250** -0.421** (0.065) (0.104) (0.180) Log of earnings 0.034** 0.139** 0.063** 0.313*** 0.115** 0.535*** (0.016) (0.063) (0.027) (0.101) (0.052) (0.174) Controls Yes Yes Yes Yes Yes Yes GP fixed effects Yes Yes Yes Yes Yes Yes N 739 739 739 739 739 739
Local Polynomial Regression Plots 8 Log of expenditure on private coaching 6 4 2 0 1 2 3 4 5 Log of total days worked under MGNREGA 95% CI kernel = epanechnikov, degree = 0, bandwidth = .5, pwidth = .61
8 Log of expenditure on private coaching 6 4 2 0 5 6 7 8 9 10 Log of earnings received from MGNREGA 95% CI kernel = epanechnikov, degree = 0, bandwidth = 1.25, pwidth = 1.87
Qualitative Evidence An agricultural labourer from the Dawaguri GP Private coaching popular in Dawaguri GP- about 70% of the children attends it Households from GPs in the haldibari block
Conclusion Evidence of impact of MGNREGA in increasing the expenditure on education through private coaching Large welfare impacts through MGNREGA Glaring problems making MGNREGA less attractive (Mukhopadhyay et al., 2015, Narayanan et al. 2016) Need to address problems related to rationing of jobs and delayed payments Higher allocation of resources to the programme
