I was reading this paper: https://ibs.org.pl/app/uploads/2015/02/IBS_Working_Paper_07_2015.pdf

The authors want to calculate employment effects following the introduction and raise of the minimum wage. They use a matching approach in order to create a treatment group (people working with less wage at period t than the minimum wage requires in t+1) and a control group (people working in t with wage 100%-130% of the minimum wage in t+1). Then, they use a DiD to check whether both groups develop the same way.

They use data on individual level!

I was wondering how they apply the DiD technically.

I think I got the idea, if they want to calculate the effect on hours worked per week. For this, they have individual data on which they can apply the DiD regression.

I do NOT get the idea if they want to calculate the effect on employment rate (in percentage points) or the share of permanent workers (look at Table 2 in the link above). How do they do it? What is the estimation model for the DiD? Do they calculate the employment rate of every year and control/treatment group respectively and use this as the dependent variable? If so, they would have just a few observations and lose a lot of information, right? Or is there any possibility to calculate the share with the DiD approach by using data on the individual level?

Hope you guys know what I mean and you give me a hint :)



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.