Page 67 - IJPS-1-1

Raphael J. Nawrotzki, Fernando Riosmena, Lori M. Hunter, and Daniel M. Runfola

land in Mexico was irrigated in 2000 (Carr, Lopez and Bilsborrow, 2009). As such, we
assume an agricultural pathway in which climate change impacts agricultural production,
leading to livelihood instabilities (Black, Adger, Arnell et al., 2011a).
In response to livelihood uncertainties, households may employ migration as a house-
hold-level risk management strategy (Massey, Arango, Hugo et al., 1993). A household
may send a migrant to an international destination to access a stable income stream
through remittances, which is independent of the local climate and market conditions
(Stark and Bloom, 1985). A number of studies have explored the relationship between
climate and migration from Mexico and found a significant relationship between rainfall
decline and international outmigration, largely from rural areas with established transna-
tional networks (Feng and Oppenheimer, 2012; Hunter, Murray and Riosmena, 2013;
Nawrotzki, Riosmena and Hunter, 2013). However, no studies have been done to investi-
gate whether climate change is associated with undocumented versus documented/legal
migrations.
Insights from related literatures suggest that climate change may influence undocu-
mented migrations in different ways than documented migrations. If climatic shocks such
as droughts impair the livelihoods of rural farmers, households may not have sufficient
time for visa applications, a process which could take years to complete (Papademetrious
and Terrazas, 2009) and would instead choose the more rapid path of undocumented bor-
der crossing. This assumption is in line with the literature on migratory responses to the
impact of economic recessions. Historical evidences suggest that economic crises in Mex-
ico have resulted in surges of undocumented migration to the U.S. (Hanson and Spili-
mbergo, 1999). Likewise, unauthorized movement is also much more responsive to eco-
nomic crises in the U.S. than movement through legal immigration channels (Papade-
metrious and Terrazas, 2009). In a similar way, climatic shocks may indirectly influence
migration dynamics through its impacts on various economic sectors (Boyd and Ibarraran,
2009) and therefore disproportionately drive undocumented migrations. Shedding some
light on this unsolved puzzle, this paper investigated whether climate change and variabil-
ity more strongly influences undocumented versus documented migrations from rural
Mexico to the U.S.

2. Data and Methods

2.1 Data
We combined detailed migration histories from the Mexican Migration Project (MMP)
(Massey, 1987) with daily temperature and precipitation information obtained from the
Global Historical Climate Network (GHCN) (Menne, Durre, Vose et al., 2012) from 214
weather stations across Mexico. Both data sets undergo rigorous quality checks and have
been used in a wide range of published research (Alexander, Zhang, Petersen et al., 2006;
Hunter, Murray and Riosmena, 2013; Massey, Durand and Pren, 2015; Wu, 2015). The
MMP started collecting data in 1982 and selects between two and five communities each
year, interviewing a random sample of 200 households in each community (Massey, 1987).
For this study, we employed data from MMP waves 1987–2013, resulting in an analytical
sample of 7,062 households located in 68 rural municipalities. Although not strictly na-
tionally representative, validation exercises have demonstrated that the MMP very accu-
rately reflects the characteristics and behavior of international migrants (Massey and
Capoferro, 2004).
The MMP data contains a wealth of sociodemographic information on all household
members and most importantly for this study, about the year of the first move to the U.S.
and the documentation status during that particular trip. This retrospective information on

International Journal of Population Studies | 2015, Volume 1, Issue 1 61