The Employment Tax Incentive (ETI) is a South African wage subsidy programme introduced in 2014 as a form of tax relief for firms hiring workers between the ages of 16 and 29, who earn less than R6 000 per month. Designed as a tool to combat high levels of youth unemployment, the results of studies estimating the effect of the ETI using survey data have shown little to no effect of the subsidy. This paper makes use of the administrative tax data made available through National Treasury in collaboration with the South African Revenue Service (SARS) to accurately identify and estimate the impact of the ETI using individual and firm-level tax returns for the period 2013 to 2016 using a Difference-in-Differences methodology combined with propensity score matching. The impact of the ETI is found to be statistically significant but small in magnitude: During a time when employment levels were decreasing, it is estimated that for every 1 job lost in a non-ETI claiming firm, ETI firms only lost between 0.51 and 0.66 jobs on average. This translates to a total of 35 333 jobs saved between 2014 and 2016 as a result of the ETI. Small firms of fewer than 10 employees have experienced the most benefit from the ETI, with growth of between 0.888 and 0.928 percentage points greater than comparable non-ETI firms, again illustrating a small, but statistically significant, effect of the policy. Furthermore, the effect of the ETI seems to be declining over time, with growth of ETI firms relative to non-ETI firms slowing down in 2015/16 relative to 2014/15, although this is not generalisable to all firm sizes. The ETI does not appear to have negatively impacted employment for workers who are thought to have been most at-risk of displacement due to the subsidy, and has not had any measurable impact on the non-wage benefits of those employed as a result of the subsidy.