The goal of this work is to find the impact of having a mother working in a ready-made garmentfactory (RMG) on schooling and work behavior of Indigenous children in Guatemala. After thesigning of...

Contact me for dta file


The goal of this work is to find the impact of having a mother working in a ready-made garment factory (RMG) on schooling and work behavior of Indigenous children in Guatemala. After the signing of the Central American Free Trade Agreement (CAFTA) in 2006, employment in these factories expanded rapidly. International investors located new factory sites in places where there would be a large number of potential employees. In practice, this meant that new factories were placed in rural locations with high levels of unemployment and low levels of literacy. The inhabitants of such locations were primarily Indigenous Guatemalans. This group comprises approximately 40% of the Guatemalan population. The 2002 and 2018 censuses of Guatemala contain detailed information on the schooling and work activities of all household members. The resulting data set contains more than 5 million observations, far too many to quickly estimate two staged least squares (2ss) intstrumental variables models. These data have been combined and a 10% random sample taken, by setting the random number seed and generating a 10 percent sample in STATA. The resulting data set is IRP2_census2002_2018_guatemala.dta. First, look for 3-4 bibliographical sources which examine the impact of RMG sectors on schooling and household resource allocation in different context. In the Estimation section, focus on indigenous Guatemalans. Some code to help with this section is contained in IRP2_FALL2022_Guatemala_RMG_schooling.do. You will need to code the summary statistics section yourself, using the example code on Courselink. Use instrumental variables techniques and fixed effects at the municipal level. The variable factory isa dummy which takes the value 1a municipality hosted a RMG factory in that year. You will need to make sure that you xtset your data so that municipal fixed effects are included but not shown. This is like the ‘areq’ command in STATA, which allows you to control for fixed effects in the cross-section without seeing the coefficient values for each regional grouping. We cannot use individual fixed effects because there is no easy way to link persons across the two years of this census. You will need to show that your estimation strategy satisfied the exclusion restriction for a good instrumental variable. The instrument should not plausibly be correlated with unobservables. That is, the instrument should affect the probability of a mother’s employment in an RMG factory only through the fact that the proximity of a person to a factory increases the probability that a person works there. Show also that the IV strategy does not work for non- Indigenous sample members (indig==0), and explain why this is the case.