Volume 43, Number 171,

October-December 2012

October-December 2012

Maquiladora Factories and Household

Income in Yucatan

Income in Yucatan

Evaluating the Problem and the *Propensity Score Matching* Method

An important question when evaluating the impacts of income on the individuals that work or do not work for an mei is to specify the average treatment effect (ate). Rosenbaum and Rubin (1983) define the ate () in a counterfactual framework as:

(3) |

Where and denote the income of individual *i* working for an mei and the individual not working for an mei, respectively. To estimate the impact of equation (3), the problem arises due to the fact that for an individual, we normally only know or , but not for both individuals. What we normally observe can be expressed as:

(4) |

Denoting *P* as the probability of observing an individual with *D = 1*, the average treatment effect (ate), , can be specified as:

(5) |

Equation (5) implies that the effect of working for an mei for the entire sample is the average weight of the effect of working for all employees in the mei (treatments) and those that do not (controls), with each weight by its relative frequency. The main problem for this causal inference comes from the unobservable counterfactuals, as and cannot be estimated (Smith and Todd, 2005).

Specifically, when available data does not provide information on the counterfactual situation, the problem of missing information increases and requires direct estimate of the effects of income from working at an mei. This study addresses this problem using the *Propensity Score Matching* (psm) method, which summarizes the characteristics of the pre-treatments for each objective within a variable index. Then it uses the Propensity Score to match with the most similar individuals possible (Rosenbaum and Rubin, 1983). The psm, which is the probability of being assigned to a conditioned treatment over pre-treatment variables, is given by:

(6) |

Where *F{.}* represents a normal cumulative distribution or logistic and X is a pre-treatment characteristic vector.^{3}

To estimate the effects of treatment based on the psm model, two important assumptions must be made. The first is the Conditional Independence Assumption (cia) mentioned above. The second is the average effect of treatment on the treated (att), defined only within the common support region. These assumptions ensure that individuals with the same values of the characteristics vector X have a positive probability of either being participants or not (Heckman *et al.*, 1997).

^{3 } The psm method controls “self-selection” by creating a counterfactual for the group of individuals working at an mei. The psm builds a statistical comparison group using the group of individuals that work and those that do not work at a maquiladora factory. In this way, the psm seeks to build an experimental database of information, conditioned over observable characteristics, resulting in a process of random selection.

Problemas del Desarrollo. Revista Latinoamericana de Economía by Instituto de Investigaciones Económicas, Universidad Nacional Autónoma de México

is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License

Published in Mexico, 2012-2018 © D.R. Universidad Nacional Autónoma
de México (UNAM).

PROBLEMAS DEL DESARROLLO. REVISTA LATINOAMERICANA DE ECONOMÍA, Volume 49, Number 194 July-September 2018 is a quarterly publication by the Universidad
Nacional Autónoma de México, Ciudad Universitaria, Coyoacán,
CP 04510, México, D.F. by Instituto de Investigaciones Económicas,
Circuito Mario de la Cueva, Ciudad Universitaria, Coyoacán,

CP 04510,
México, D.F. Tel (52 55) 56 23 01 05 and (52 55) 56 24 23 39, fax
(52 55) 56 23 00 97, www.probdes.iiec.unam.mx, revprode@unam.mx. Journal
Editor: Moritz Cruz. Reservation of rights
to exclusive use of the title: 04-2012-070613560300-203, ISSN: pending.
Person responsible for the latest update of this issue: Minerva García,
Circuito Maestro Mario de la Cueva s/n, Ciudad Universitaria, Coyoacán,
CP 04510, México D.F., latest update: August 29^{th}, 2018.

The opinions expressed by authors do not necessarily reflect those of the
editor of the publication.

Permission to reproduce all or part of the published texts is granted provided the
source is cited in full including the web address.

Credits | Contact

The online journal *Problemas del Desarrollo. Revista Latinoamericana
de Economía* corresponds to the printed edition of the same title with
ISSN 0301-7036