returns to computer use in bangladesh: an …...the gaps in the labour market. computer skills,...
TRANSCRIPT
Fahmida Khatun1
Syed Yusuf Saadat2
ABSTRACT
In the highly competitivelabour market of Bangladesh,workers are unable to find theright jobs and employers arealso unable to find the rightworkers. Such skills mismatchis a manifestation thateducation alone cannot bridgethe gaps in the labour market.Computer skills, which cancomplement education, arebecoming increasinglyimportant. Nevertheless, thereis a dearth of researchregarding the potential benefitof possessing computer skills inthe labour market ofBangladesh. This paper intendsto fill in this knowledge gap byestimating the returns tocomputer use using a nationallabour force survey dataset. Byutilizing Heckman’s two-stepselection model, it was foundthat workers who could usecomputers earned 17 per centmore than workers who couldnot use computers.Additionally, the returns toeducation and experience wereestimated to be 3 per cent and 2per cent respectively. Thisimplies that computer skillswere as valuable as 5 years ofeducation or 8 years ofexperience, assuming that thereturns to all three wereconstant over time. The resultsof this study provide empiricalevidence in favor of allocatinggovernment resources forcomputer training, and alsoadvocate for individualinvestment towards learningcomputer skills.
THE PROBLEMATIQUE
cpdbd
1 Executive Director, Centre for Policy Dialogue (CPD)2 Research Associate, Centre for Policy Dialogue (CPD
RETURNS TO COMPUTER USE IN BANGLADESH: AN ECONOMETRIC ANALYSIS
cpd.org.bd CPDBangladesh
Centre for Policy Dialogue (CPD)
syedyusufsaadat syedyusufsaadat
Syed Yusuf Saadat
96
67
4
33
020406080
100
Do not have computer Have computer
Per
cent
Computer ownership and computer use (in percentage)
Never used computer Used computer
Wages
Region
Never used computer (in BDT)
Used computer (in BDT)
Wage differential (in %)
Barishal 11,343 15,731 38.68Chattogram 11,217 14,358 28.00Dhaka 10,881 14,788 35.91Khulna 10,709 13,122 22.53Rajshahi 10,620 13,425 26.41Rangpur 10,264 13,551 32.02Sylhet 10,320 13,401 29.85Rural 10,287 12,925 25.64Urban 11,254 14,434 28.26National 10,812 14,162 30.98
010203040506070
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
2015
2017
Per
cent
Year
Sector-wise employment (as percentage of total employment) in Bangladesh (1991-2017)
Agriculture Industry Services
Rural Agricultural
Sector Urban Industrial
Sector
Rural Agricultural
Sector
Urban Industrial
Sector
Less DevelopedEconomy
DevelopedEconomyStructural Change
Market wage model specification
ln (𝑊&) = 𝑏* + 𝑏,𝑆& + 𝑏.𝐸& +𝑏0 𝐸&. +𝑏1 𝐶& + 𝑢& (𝑖)
Education
WR
WM(est.)
WM(true)
OBSERVED
UNOBSERVED
WM
Sample selection bias
Labor force participation model specification
𝑙𝑛(𝑊&∗) = 𝛽* + 𝛽,ℎ& + 𝛽.𝐴𝑖 + 𝛽0𝑀& + 𝛽1𝐾& + 𝛽=𝑃& + 𝜀&(𝑖𝑖)
Selection bias corrected model specification
𝑙𝑛(𝑊&) = 𝑏* + 𝑏,𝑆& + 𝑏.𝐸& + 𝑏0𝐸&. + 𝑏1𝐶& + 𝑏=𝜆& + 𝑢&(iii)
THE RESULTSModel without computer Model with computer
Regression Probit Heckman Probit HeckmanVariable lnwage lnwage lnwage lnwageeducation 0.0343*** 0.0302***
(0.0010) (0.0010)experience 0.0198*** 0.0203***
(0.0007) (0.0007)experience2 -0.0002*** -0.0002***
(0.0000) (0.0000)computer 0.1702***
(0.0123)hours 0.0107*** 0 .0107***
(0.0005) (0.0005)assets -0.0003*** -0.0003***
(0.0000) (0.0000)married -0.3196*** -0.3196***
(0.0171) (0.0171)children -0.0234** -0.0234**
(0.0096) (0.0096)CPI -0.0415*** 6.9558***
(0.0024) (0.4395)lambda -0.1805 -0.1558
(0.0128) (0.0133)Constant 6.9558*** 7.4804*** 6.9558*** 7.4678***
(0.4395) (0.0181) (0.4395) (0.0182)LR chi2 1730.34 1730.34Wald chi2 1794.51 2030.21Prob > chi2 0.0000 0.0000 0.0000 0.0000
THE APPROACH
0
20
40
60
80
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Per
cent
Years
Returns to education, experience, and computers
Returns to education (cumulative)Returns to experience (cumulative)Returns to computer use
17%
https://cpd.org.bd