Estimating the Short Run Effects of South Africa's Employment Tax Incentive on Youth Employment Probabilities using A Difference‐in‐Differences Approach

• Published date: 01 June 2016
• Author: Vimal Ranchhod,Arden Finn
• DOI: http://doi.org/10.1111/saje.12121
• Date: 01 June 2016

ESTIMATING THE SHORT RUN EFFECTS OF SOUTH

AFRICA’S EMPLOYMENT TAX INCENTIVE ON YOUTH

EMPLOYMENT PROBABILITIES USING A DIFFERENCE-IN-

DIFFERENCES APPROACH

VIMAL RANCHHOD*AND ARDEN FINN

†

Abstract

South Africa’s Employment Tax Incentive (ETI) came into effect on the 1st of January 2014,

with the objective of reducing the substantial national youth unemployment rate. Under the

ETI, firms are eligible to claim a deduction from their taxes due, for the portion of their wage

bill that is paid to certain groups of youth employees. We utilise several waves of nationally repre-

sentative data and implement a difference-in-differences methodology at the individual level, in

order to identify the effects of the ETI on youth employment probabilities in the short run. Our

primary finding is that the ETI did not have any statistically significant and positive effects on

youth employment probabilities. The point estimate from our preferred regression is 20.005

and the 95% confidence interval is from 20.017 to 0.006. We also find no evidence that the

ETI has resulted in an increase in the level of churning in the labour market for youth. Thus,

any decrease in tax revenues that arise from the ETI are effectively accruing to firms which, col-

lectively, would have employed as many youth even in the absence of the ETI.

JEL Classification: H25, H32, J38

Keywords: Youth, unemployment, South Africa, wage subsidy, employment tax incentive

1. INTRODUCTION

Stubbornly high levels of unemployment in general, and extremely high levels of youth

unemployment in particular, have been constant features of South African society since

the transition to democracy in 1994. The youth (defined in this paper as adults aged

between 18 and 29) experience an unemployment rate that is more than double that of

the overall population unemployment rate, and about two-thirds of unemployed youth

have never had a job (National Treasury, 2011). Policy discussions aimed at boosting

youth employment have been taking place since the mid-2000s, with the objective of

* Corresponding author: Chief Research Office, SALDRU, School of Economics, University

of Cape Town, Economics, Room 7.61, PD Hahn Bldg, Upper Campus, Rondebosch,

Western Cape, South Africa 7700. E-mail: Vimal.ranchhod@uct.ac.za

†

Southern Africa Labour and Development Research Unit, University of Cape Town

The authors wish to thank Ihsaan Bassier for excellent research assistance, and participants in

the SALDRU seminar series for very useful comments and suggestions. All opinions, errors

and omissions expressed in this paper remain the sole responsibility of the authors. Ranchhod

and Finn acknowledge support from the National Research Foundation’s Human and Social

Dynamics in Development Grand Challenge.

V

C2016 Economic Society of South Africa. doi: 10.1111/saje.12121

199

South African Journal of Economics Vol. 84:2 June 2016

South African Journal

of Economics

raising employment levels and mitigating the “wasting” effect of being unemployed for

an extended period of time. These discussions led to the Employment Tax Incentive

(ETI), which was initiated on the 1st of January 2014. The ETI takes the form of a

direct intervention in the labour market to reduce the cost to firms of employing young

workers. This is one of the most ambitious youth employment initiatives in the country’s

history, with a proposed cost to government of R5 billion over 3 years intended to create

178,000 new jobs for youth over that period.

In this paper, we ask the question “Did the youth employment tax incentive have an

impact on youth employment rates in South Africa in the short run?” We utilise data

from the nationally representative Quarterly Labour Force Survey (QLFS) from 12 quar-

ters before and 2 quarters after the implementation of the ETI in order to identify the

short run effects of the intervention. We employ a number of difference-in-differences

estimators with different control groups to identify what effect the programme had on

the probability of youth employment. We find no evidence of a positive effect on youth

employment probabilities in the short run, and no evidence that the implementation of

the ETI has led to increased levels of churning amongst youth in the labour market.

The remainder of this paper is structured as follows. Section 2 presents a background

to the ETI in South Africa and summarises both local and international evidence on

youth employment incentives. In Section 3, we discuss the rules of the ETI. Section 4

describes the datasets and variables that we use, while Section 5 outlines the methods

that we use. Section 6 presents the main results of our analysis, Section 7 discusses the

outcomes of some robustness checks and Section 8 provides some concluding discussion.

2. BACKGROUND TO THE ETI IN SOUTH AFRICA AND INTERNATIONAL

EVIDENCE

One of the first papers to discuss a youth wage subsidy for South Africa is by Levinsohn

(2007). He argues that when firms hire workers, they predict a productivity level based

on training and experience. For youth in South Africa, education is a weak signal and in

addition, young people have a shortfall of work experience. Moreover, firms face substan-

tial costs in having to train inexperienced workers. Firms may thus prefer to hire older

workers who can signal their productivity through experience.

Thus, most interventions have taken the form of one of two strategies. The first

branch aims to address the lack of relevant skills or reliable education by raising the qual-

ity of workers through supply-side interventions. The second branch offsets the cost

(thereby lowering the risk) of employing inexperienced/young workers through demand-

side initiatives.

The ETI is a demand-side intervention. In South Africa, there have been two major

demand-side interventions; the Expanded Public Works Programme (EPWP) and Lear-

nership Agreements.

1

The EPWP provided 1.6 million jobs in its first phase, but is

widely recognised as a short-term and unsustainable solution (Meth, 2011 and Betcher-

man et al., 2004). A Learnership is a policy whereby firms are subsidised for providing

approved training to employees, integrated with a job. While the effects of Learnerships

have not been established, Schoer and Rankin (2011) found that just 19% of firms offer

1

As defined in Section 17 of the Skills Development Act.

200 South African Journal of Economics Vol. 84:2 June 2016

V

C2016 Economic Society of South Africa.