Page 74 - IJPS-1-1

Undocumented migration in response to climate change

Table 2. Multi-level discrete-time event history models predicting the odds of undocumented and documented international migrations from rural
Mexico, 1986–1999
Undocumented Documented
b sig. b sig.
Household level (head)
Female 0.53 *** 0.68
Married 0.96 1.36
No. of children 0.90 ** 0.99
Education 0.74 ** 3.29 ***
a
Working experience 0.71 *** 1.00
a
Occupation: not in labor force 0.91 1.45
Occupation: white collar 0.50 *** 0.63
Owns property 0.83 * 1.14
Owns business 0.77 * 1.03
Community/municipality level
a
Network density 1.56 *** 1.49 **
Wealth index 0.81 0.74
Corn (area harvested) 0.94 0.68 *
a
Farmland irrigated 1.04 0.93
Base period precip (1961-90) 1.11 0.73
Base period temp (1961-90) 0.91 ** 0.98
a
Male labor in agriculture 1.01 0.88
Climate change
Warm spell duration 1.19 *** 1.16
Precip extremely wet days 0.82 *** 0.98
Model statistics
Var. Intercept (Mun) 0.215 0.718
BIC 8451 1703
N (HH-year) 67511 67511
N (HH) 7062 7062
N (Mun) 68 68
a
Notes: Coefficients reflect odd ratios; Coefficients relate to an incremental change of 10 units; baseline hazard of migration was included as a multi-part intercept
using year dummies (not shown); all models control for the survey year to account for recall bias (not shown); Occupation: Blue collar used as reference; all predictors
were lagged by one year; low values on the Variance Inflation Factor (VIF) demonstrated that multi-collinearity does not bias the estimates; a jack-knife type procedure
was performed, iteratively removing one municipality from the sample and re-estimating the model (Nawrotzki, 2012; Ruiter & De Graaf, 2006). The results showed
that the estimates for the climate change predictors are highly robust;
* p < 0.05; ** p < 0.01; *** p < 0.001

crossings to stabilize their livelihoods and access alternative income streams through re-
mittances.
The directionality of significant climate change effects suggests a rise in undocumented
international migrations in response to a warming in temperatures. Heat waves and tem-
perature increases are problematic for the agricultural sector and are associated with a de-
cline in crop yield (Lobell, Hammer, McLean et al., 2013). Adverse impacts on agricul-
tural productivity may lead to a decline in income and employment opportunities to which
households may respond with increased levels of migration (Bohra-Mishra, Oppenheimer
and Hsiang, 2014; Mueller, Gray and Kosec, 2014).
In contrast, increases in precipitation led to a decline in undocumented migrations. Only

International Journal of Population Studies | 2015, Volume 1, Issue 1 68