Droughts and domestic violence: Measuring the gender-climate nexus

Aguila -Gómez, Sand a; Salaza -Díaz, And ea
Wo king Pape
D ough s and domes ic iolence: Measu ing he gende -
clima e nexus
IDB Wo king Pape Se ies, No. IDB-WP-1653
P o ided in Coope a ion wi h:
In e -Ame ican De elopmen Bank (IDB), Washing on, DC
Sugges ed Ci a ion: Aguila -Gómez, Sand a; Salaza -Díaz, And ea (2025) : D ough s and domes ic
iolence: Measu ing he gende -clima e nexus, IDB Wo king Pape Se ies, No. IDB-WP-1653, In e -
Ame ican De elopmen Bank (IDB), Washing on, DC,
h ps://doi.o g/10.18235/0013368
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/309193
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/3.0/igo/
D ough s and Domes ic
V
iolence:
Measu ing he Gende -Clima e Nexus
Sand a Aguila -Gómez
And ea Salaza -Díaz
WORKING PAPER No IDB-WP-1653
In e -
A
me ican De elopmen Bank
Gende and Di e si y Di ision
Janua y 2025
* Uni e sidad de los Andes
** Kennedy School o Go e nmen , Ha a d Uni e si y
D ough s and Domes ic
V
iolence:
Measu ing he Gende -Clima e Nexus
Sand a Aguila -Gómez*
And ea Salaza -Díaz**
In e -
A
me ican De elopmen Bank
Gende and Di e si y Di ision
Janua y 2025
Ca aloging-in-Publica ion da a p o ided by he
In e -Ame ican De elopmen Bank
Felipe He e a Lib a y
Aguila Gómez, Sand a.
D ough s and domes ic iolence: measu ing he gende -clima e nexus /
Sand a Aguila -Gomez, And ea Salaza -Diaz.
p. cm. — (IDB Wo king Pape Se ies ; 1653)
Includes bibliog aphic e e ences.
1. In ima e pa ne iolence-Mexico. 2. Women-Violence agains -Mexico. 3.
Clima ic changes-Social aspec s-Mexico. 4. Gende mains eaming-Mexico. I.
Salaza -Diaz, And ea. II. In e -Ame ican De elopmen Bank. Gende and
Di e si y Di ision. III. Ti le. IV. Se ies.
IDB-WP-1653
h p://www.iadb.o g
Copy igh © 2025 In e -Ame ican De elopmen Bank ("IDB"). This wo k is subjec o a C ea i e
Commons license CC BY 3.0 IGO (h ps://c ea i ecommons.o g/licenses/by/3.0/igo/legalcode). The
e ms and condi ions indica ed in he URL link mus be me and he espec i e ecogni ion mus be
g an ed o he IDB.
Fu he o sec ion 8 o he abo e license, any media ion ela ing o dispu es a ising unde such license
shall be conduc ed in acco dance wi h he WIPO Media ion Rules. Any dispu e ela ed o he use o
he wo ks o he IDB ha canno be se led amicably shall be submi ed o a bi a ion pu suan o he
Uni ed Na ions Commission on In e na ional T ade Law (UNCITRAL) ules. The use o he IDB's name
o any pu pose o he han o a ibu ion, and he use o IDB's logo shall be subjec o a sepa a e
w i en license ag eemen be ween he IDB and he use and is no au ho ized as pa o his license.
No e ha he URL link includes e ms and condi ions ha a e an in eg al pa o his license.
The opinions exp essed in his wo k a e hose o he au ho s and do no necessa ily e lec he iews o
he In e -Ame ican De elopmen Bank, i s Boa d o Di ec o s, o he coun ies hey ep esen .
Abs ac *
E e y yea , 245 million women a e ic ims o in ima e pa ne iolence (IPV). Clima e change is
hypo hesized o exace ba e his igu e h ough i s dis up i e impac on household li elihoods, among
o he channels. Howe e , li le causal e idence exis s on his aspec o he clima e-gende nexus,
pa ly due o measu emen challenges ha ha e con ibu ed o gaps in he li e a u e. In his pape ,
we use h ee di e en IPV da a sou ces o examine he e ec o d ough in Mexico and he ole o
ag icul u al ulne abili y in in ensi ying hese e ec s. We ind obus e idence o inc eases in all
measu es o IPV in esponse o local p ecipi a ion de ici s: as unan icipa ed exposu e o days
wi hou ain in he p e ious mon h ises, mo e inju ies linked o IPV a e eco ded in he public
heal h sys em, police epo s inc ease, and mo e 911 calls ela ed o IPV a e made. The e ec s a e
s onge in egions highly dependen on ag icul u e, pa icula ly when he shock occu s du ing he
g owing season. In a coun y whe e mos ag icul u al income and land a e con olled by men, ou
esul s align wi h heo e ical p edic ions om male-backlash IPV models and ex ac i e iolence
models. We also ind ha he impac o d ough on IPV is mo e p onounced in municipali ies
wi h low s a e capaci y, hough po en ial di e ences in epo ing beha io be ween IPV measu es
complica e compa isons. Ou indings unde sco e he need o design gende -sensi i e disas e elie
policies, s eng hen us in epo ing mechanisms and helplines, and educe he social accep abili y
o IPV.
JEL classi ica ions: I15, J16, Q54, D13
Keywo ds: D ough , Domes ic iolence, In ima e pa ne iolence, Ag icul u e,
Mexico
∗This p ojec was pa ially unded by he In e -Ame ican De elopmen Banks (IDB) Gende and Di e si y Lab.
We hank he semina pa icipan s o his p og am o hei help ul ques ions and eedback. We a e also g a e ul
o And és Moya and Ambe Pe e man o hei commen s and sugges ions and o he IDB GDLab eam o hei
con inued suppo : Ka en Ma ínez, Monse a Bus elo, and Claudia Ma ínez. We hank Da id Chapa o-Ál a ez,
Angel Ramí ez and Valen ina Cas illa-Gu ié ez o hei esea ch assis ance.

1 In oduc ion
In ima e pa ne iolence (IPV) is one o he mos p essing challenges o women’s heal h wo ldwide:
app oxima ely 245 million women—equa ing o 10% o hose aged 15 and abo e—a e subjec ed o
ei he psychological, sexual, o physical iolence by hei pa ne s e e y yea . In Mexico, he ocus
o ou s udy, he p e alence o IPV e lec s global pa e ns. Acco ding o he Wo ld Bank Gende
Po al, 24% o Mexican women ha e expe ienced IPV, closely aligning wi h he global a e age o
26% (Wo ld Bank,2022b).
Despi e his high p e alence, epo ing a es a e low: only one in 10 women epo ed expe i-
encing iolence in he pas 12 mon hs (ENDIREH,2021). Thus, e o s o imp o e policies aimed
a p o ec ing women a e hampe ed by a lack o high-quali y da a. Fu he mo e sys ema ic da a
collec ion is ecen in many egions wi h high p e alence (Cullen,2023). The e ec s o IPV each
beyond immedia e physical o emo ional ha m, signi ican ly in luencing long- e m heal h ou comes
(Campbell,2002), educing labo ma ke pa icipa ion (Bha acha ya,2015;Sabia e al.,2013),
and diminishing o e all well-being (Odu o e al.,2015;Adams e al.,2012).
IPV has been poised as a key dimension o he clima e-gende nexus. Clima e change, which
is an icipa ed o escala e he equency and se e i y o ex eme en i onmen al e en s (A ias e al.,
2021), in e ac s wi h exis ing inequali ies (Kee hi a ne and Tol,2018;Wa and Aung,2019;
Chowdhu y e al.,2021)), including gende (Ge e s e al.,2020;UN Women,2022), disp opo ion-
a ely a ec ing women’s mo ali y (Neumaye and Plümpe ,2007), mo bidi y (Fa ema e al.,2019),
and housing secu i y (Logie e al.,2017). 1An en i onmen al haza d o pa icula conce n a e
d ough s, which a e among he mos expensi e na u al disas e s in he wo ld, wi h signi ican cos s
o ecosys ems, ag icul u e, and human li elihoods (Cook e al.,2018). Clima e change is expec ed
o inc ease d ough equency and se e i y o e much o he globe (Dai,2013). By 2050, an addi-
ional 1 billion people a e expec ed o li e wi h ex emely high wa e s ess2(Kuzma e al.,2023).
This pape seeks o unde s and he causal impac o d ough on IPV in Mexico, a coun y wi h
bo h opical and semi-a id en i onmen s, whe e 40-80% o he popula ion a e exposed annually
o wa e s ess (Kuzma e al.,2023).
Ag icul u e is a pa icula ly clima e-sensi i e sec o , and in Mexico, i is also p edominan ly
male-biased in e ms o labo composi ion, as well as land owne ship and go e nance. Women
cons i u e only 12.7% o he coun y’s ag icul u al labo e s. Fu he mo e, while hal o he land
is unde a communal p ope y and managemen sys em known as ejidos, women hold only 26% o
ejido igh s, and ou o he 14.6 housand ejidos and communi ies wi h ep esen a i e bodies, only
1Some ways in which gende no ms educe women’s esilience include hei lowe con ol o e income and asse s
o ace shocks (Van Aels and Hol oe ,2016;Roncoli e al.,2001) and acing mobili y es ic ions ha limi hei
access o aid and heal hca e (Pa ma e al.,2019;Boye e al.,2020)
2Wa e s ess, he a io o wa e demand o enewable supply, measu es he compe i ion o e local wa e esou ces.
The smalle he gap be ween supply and demand, he mo e ulne able a place is o wa e sho ages.
1
7.4% a e p esided o e by a woman (Regis o Ag a io Nacional,2019). Unde hese condi ions,
an a e age dec ease in c op yields esul ing om d ough s di ec ly educes he ela i e income o
men compa ed o women. Economic heo ies o IPV p o ide di e ing p edic ions on he impac
o disas e s on IPV. Ba gaining powe models sugges ha a dec ease in men’s ela i e income
should inc ease women’s ba gaining powe wi hin he household and consequen ly dec ease domes ic
iolence (Aize ,2010;Ande be g e al.,2016). On he o he hand, ex ac i e iolence models a gue
ha educed income may inc ease male pa ne s’ incen i es o use iolence as a ool o egain
con ol o e household decision-making o ex ac inancial esou ces, o compel women o wo k
longe hou s, o as a esponse o s ess o e income loss (Eswa an and Malho a,2011;Bobonis
e al.,2013;E en and Keskin,2018,2021,2024).
In ou s udy, we in es iga e he impac s o d ough exposu e on IPV in Mexican munici-
pali ies. We combine g idded p ecipi a ion, and empe a u e es ima es da a om NASA Dayme
eposi o y (Tho n on e al.,2022) o gene a e disas e exposu e measu es wi h high empo al and
spa ial g anula i y wi h adminis a i e da a om a ious sou ces. The iden i ying assump ion is
ha hese e en s a e unexpec ed o households once con olling o a ba e y o geog aphical and
seasonal ixed e ec s and hus plausibly exogenous. To o e come he measu emen challenge, we
employ h ee measu es o IPV wi h dis inc s eng hs and weaknesses. Fi s , we u ilize hospi al
discha ge eco ds epo ed by heal hca e ins i u ions om Mexico’s la ges public heal h subsys-
em. We agg ega e hese da a a he weekly le el o p oduce municipali y-week measu es o he
incidence o inju ies ha can be disagg ega ed in o IPV and nondomes ic iolence. Second, we
inco po a e police iles om he Na ional Secu i y Sys em (SNSP) o domes ic iolence da a. Fi-
nally, h ough hund eds o in o ma ion eques s o local law en o cemen agencies, we collec ed
da a om local go e nmen ho lines ha ecei e eme gency calls abou IPV. O e all, wo ou o
h ee o ou gende -based iolence (GBV) measu es di ec ly cap u e IPV. Al hough we can only
measu e domes ic iolence in police iles wi hou di e en ia ing by sex, public hospi al eco ds show
ha 82% o domes ic iolence is IPV, and 95% o IPV ic ims in Mexico a e women. The e o e,
h oughou he documen , we will p ima ily e e o ou esul s as IPV. Wi h he h ee men ioned
measu es, we conduc ou analyses a he weekly and mon hly le els and quan i y he e ec on
iolence o each addi ional ex eme day in a gi en week.
Ou indings e eal a obus causal e ec o d ough on IPV and non-domes ic iolence
owa ds women and non-domes ic iolence owa ds men. The a e o emale hospi aliza ions o
non-domes ic iolence inc eases by 7.4%, and o male hospi aliza ions, he inc ease is 14%. IPV
a es display a gende ed pa e n; emale hospi aliza ions o IPV inc ease by up o 9% du ing
se e e d ough pe iods, while he e is no s a is ical impac on male IPV hospi aliza ions. Using
police epo s and 911 ho line calls as al e na i e dependen a iables, we consis en ly ind ma ked
inc eases in IPV du ing d ough s. Regions wi h lowe s a e capaci y exhibi highe inc eases in
police epo s and hospi aliza ions, sugges ing ha s a e capaci y in luences he de e ence e ec
2
o IPV. In a de e io a ion o he ule o law, unde - epo ing biases migh inc ease. Wi h his
mechanism ope a ing agains ou abili y o cap u e inc eases in c ime, de ec ing such inc eases
in low-s a e capaci y coun ies—ce e is pa ibus—sugges s ha he ise in incidence ou weighs he
possible bias in epo ing.
The se ing o ou s udy is ele an o his heo e ical deba e because d ough s do no
usually des oy in as uc u e o p ope y, he eby no a ec ing some common igge s o iolence
such as he de e io a ion o s a e capaci y, des uc ion o oads, which weakens suppo sys ems o
women, and eme gency displacemen (Thu s on e al.,2021;Ge e s e al.,2020;UN Women,2022).
This unique cha ac e is ic o d ough s allows us o isola e he economic mechanisms, pa icula ly
income loss—o s ess due o an icipa ed income loss, a mechanism especially unde sco ed in he
public heal h li e a u e (Thu s on e al.,2021)— o es exis ing heo ies o IPV. We examine he
mechanisms behind ou main indings, highligh ing he ole o ag icul u al down u ns in a coun y
whe e he bulk o he wo k o ce and land owne ship in his sec o is male. We ind ha d ough s
signi ican ly inc ease IPV in egions wi h high ag icul u al eliance. Ou esul s align mo e obus ly
wi h he "male backlash" model, whe e male s a us inconsis ency heigh ens IPV as men seek o
asse con ol wi hin he household o ex ac esou ces o compensa e o income losses. We ind
ha e ec s a e mo e p onounced when he d ough hi s du ing he municipali y- and c op-speci ic
g owing seasons, unde sco ing he ole o immedia e economic s ess once a me s obse e low
p ecipi a ion le els.
Ou wo k con ibu es o he li e a u e on IPV measu emen , which is an impo an challenge
in iden i ying i s causes. P e ious wo ks ha e u ilized su ey da a (e.g., in Díaz and Salda iaga
(2023); S ia schi and T ako (2024)), police epo s (Mille e al.,2024) hospi al eco ds (Aize ,
2010;S ia schi and T ako,2024), and ecen ly, eme gency calls (Leslie and Wilson,2020;Agüe o,
2021;Mille e al.,2024). While ecen pape s ha e highligh ed he ad an ages o ho line calls as
a measu e o iolence agains women, in his pape we discuss hei ad an ages and d awbacks by
compa ing his measu e o se e al yea s o police epo s and hospi al eco ds a he na ional le el.
We con i m ha hospi aliza ions and police epo s expe ience highe inc eases in municipali ies
wi h low local s a e capaci y. Howe e , u he esea ch is equi ed o d aw conclusions abou 911
calls. In any case, ou indings highligh he impo ance o c ea ing accessible suppo se ices
o women o seek help and epo ins ances o IPV. This includes he signi icance o go e nmen s
ga he ing and making public measu es o IPV.
Ou collec ion o da ase s enables us o analyze p edominan ly public space iolence and
compa e i wi h IPV, which mainly occu s a home. This le s us show ha , while gene alized
iolence esponds o d ough s, women ace an addi ional bu den a home. Wi h hese insigh s,
ou s udy con ibu es o a small bu g owing body o quan i a i e analyses ha in es iga e he
ela ionship be ween d ough s and gende iolence. This li e a u e has ocused on low-income
3
coun ies, equen ly inding signi ican impac s (Hi onen e al.,2023;Eps ein e al.,2020;Munala
e al.,2023a;Abiona and Koppens eine ,2018;Cools and Ko sadam,2017;Coope e al.,2021).
Howe e , middle-income coun ies, home o 75% o he wo ld’s popula ion, emain unexplo ed.
Ou indings o he s ong e ec s o d ough s on iolence in Mexico, he wel h-la ges economy
in he wo ld, sugges ha women’s ulne abili y o clima e change will no sol e i sel h ough
economic de elopmen . As clima e change un a els, mo e policies add essing he gende ed impac s
o clima e ex emes will be equi ed.3
We also con ibu e o unde s anding he clima e-gende nexus h ough disas e s. Thu s on
e al. (2021) inds in a li e a u e e iew ha quan i a i e s udies wi h a g ea e geog aphical
scope and use o spa ial da a a e equi ed o unde s and he magni ude o he p oblem and he
mechanisms behind i . They iden i y ha some au ho s o quan i a i e pape s p esen hypo hesized
pa hways om disas e exposu e o inc eased iolence agains women and gi ls (VAWG) bu ail o
es i empi ically. While some wo k has looked a he e ec s o d ough on sel - epo ed measu es
o IPV using epea ed c oss-sec ion su ey da a (Díaz and Salda iaga,2023;Cools e al.,2020),
mos o he cu en li e a u e comp ises quali a i e in es iga ions (Ylipaa e al.,2019) and case-
speci ic analyses o indi idual ex eme e en s ( an Daalen e al.,2022). Ou indings a e consis en
wi h a model in which men eac o income loss and s ess by exe ing mo e iolence owa d hei
pa ne s.
Mo e b oadly, we con ibu e o ou unde s anding o he clima e- iolence nexus. We ind
ha he e ec o d ough on iolence agains women is la ge han he clima e-con lic es ima es
epo ed in many p e ious s udies4. We es ima e an inc ease be ween 9% and 26%—depending on
he a iable employed o measu e IPV— om he sample mean when a ull mon h o unexpec ed
d ough hi s a municipali y. The magni ude o hese e ec s in a middle-income coun y like Mexico
sugges s ha he e ec s o gende ed iolence a e a c ucial, p e iously unexplo ed dimension o he
clima e- iolence li e a u e.
The emainde o his pape is o ganized as ollows. Sec ion 2p o ides a heo e ical back-
g ound and con ex , Sec ion 3desc ibes he da a used in ou analysis and p esen s desc ip i e
s a is ics. Sec ion 4lays ou ou empi ical s a egies. Sec ion 5p esen s ou baseline esul s and
analyzes he mechanisms, he e ogenei y, and obus ness checks. Sec ion 6concludes.
3Figu e A.1 shows ha Mexico’s GDP pe capi a is be ween 1.5 and 6 la ge han o he coun ies s udied in
he quan i a i e d ough -IPV li e a u e: 2.8 imes he GDP pe capi a o India, 5.3 imes he Sub-Saha an A ica
a e age, and 1.6 imes he GDP pe capi a o Pe u, he only o he La in Ame ican coun y s udied in his li e a u e.
4In a e iew o his li e a u e, Hsiang e al. (2013) ind ha he e ec o empe a u e is gene ally la ge han
he s anda dized e ec o ain all, and he e ec on in e g oup iolence (e.g., ci il wa ) is la ge han he e ec on
in e pe sonal iolence (e.g., assaul ). Pa icula ly, o each s anda d de ia ion change in clima e owa d wa me
empe a u es o mo e ex eme ain all, median es ima es indica e ha he equency o in e pe sonal iolence ises
4% and he equency o in e g oup con lic ises 14%.
4
a) 911 call da a a ailabili y by s a e
b) Co e age pe iod
S a e A ailabili y
Aguascalien es 2005-2021
Bajacali o nia, Sono a 2006-2021
Chiapas 2007-2021
Baja Cali o nia Su 2015-2021
Nue o Leon, Ve ac uz 2016-2021
Dis i o Fede al, Puebla, Sinaloa, San Luis de
Po osi, Campeche, Quin ana Roo
2017-2021
Mexico, Guanajua o, Michoacan, Gue e o, Hi-
dalgo, Tabasco, Yuca an, Naya i
2018-2021
Jalisco, Chihuahua 2019-2021
Oaxaca 2010-2021
Zaca ecas 2011-2021
Figu e 1: 911 da a a ailabili y
No e: Panel a) shows he a ailabili y o ho line da a by s a e. S a es a e so ed by popula ion size acco ding o
he 2010 Census. Panel b) Shows he speci ic yea s a ailable o each s a e ha p o ided in o ma ion h ough he
anspa ency pla o m.
3.1.4 Desc ip i e s a is ics
While nei he police eco ds no ho line da a egis e he sex o he ic im, we know om public
hospi al eco ds ha 82% o domes ic iolence is IPV and ha 95% o IPV ic ims in Mexico
a e women. The e o e, h oughou he esul s analyses, we will p ima ily e e o ou esul s as
IPV.
In ima e pa ne iolence is a highly eminized c ime and he mos common o m o iolence
11

agains women, as shown in Table 1. The able is also in o ma i e in e ms o illus a ing he
“icebe g" o domes ic iolence. E e y mon h in Mexican municipali ies, 208 eme gency calls pe
100,000 inhabi an s a e made on a e age in esponse o IPV and 67 inciden s a e epo ed o he
police. E e y week, 23 women pe 100,000 inhabi an s seek heal hca e due o iolence pe pe a ed
by hei pa ne .
Table 1: Violence da a and desc ip i e s a is ics
Panel A: Hospi al weekly eco ds a e pe 100,000 inhabi an s
N Mean SD Min Max
Female ic ims o assaul 1,003,808 2.55 11.5 0 1149
Male ic ims o assaul 1,003,808 3.0 12.74 0 658
Female ic ims o IPV 1,003,808 5.71 29.2 0 3349
Male ic ims o IPV 1,003,808 0.11 1.85 0 554
Female ic ims o child mal ea men 1,003,808 0.42 4.01 0 707
Male ic ims o child mal ea men 1,003,808 0.09 1.70 0 455
Female ic ims o sexual iolence 1,003,808 0.28 2.82 0 233
Male ic ims o sexual iolence 1,003,808 0.025 0.78 0 220
Panel B: Mon hly c ime epo s a e pe 100,000 inhabi an s
N Mean SD Min Max
Domes ic Violence 193,380 66.77 213 0 8940
Panel C: Mon hly ho line da a a e pe 100,000 inhabi an s
N Mean SD Min Max
IPV 110,174 208 1225 0 185913
No es: Panel A) Hospi al Discha ge Subsys em (SAEH) (2012-2019), weekly da a om 75% o Mexican
municipali ies. Panel B) Na ional Secu i y Sys em (SNSP) (2015-2021), mon hly da a om all Mexican
municipali ies. Panel C), Mexico’s 911 Sys em (2016-2021) Mon hly da a om municipali ies wi h a ailable
da a as desc ibed in Figu e 1, co e ing he place o esidence o 76% o he na ional popula ion.
3.2 D ough
We cons uc a measu e o exposu e ha coun s he anomalous days wi hou ain in a ious e e -
ence pe iods (e.g., pas week, pas mon h). Anomalous days a e unexpec ed days wi hou ain in
a speci ic municipali y and week o he yea . We use p ecipi a ion da a om Dayme (Tho n on
e al.,2022), a da a p oduc p o ided by NASA’s Oak Ridge Na ional Labo a o y (ORNL). This
da ase p o ides da a on a 1 km x 1 km g idded su ace o e con inen al No h Ame ica since 1980.
The es ima es a e gene a ed h ough he in e pola ion and ex apola ion o daily me eo ological
obse a ions, wi h model inpu s encompassing a digi al ele a ion model, de i ed ho izon iles, and
a land-wa e mask. The esul ing ou pu p o ides g idded es ima es o o al daily p ecipi a ion.
We selec ed his sou ce o wea he in o ma ion no only because i has a high spa ial esolu ion bu
12
also because i has he bes pe o mance o daily p ecipi a ion in Mexico (Rincón-A alos e al.,
2022).
We also p ocess he a e age daily maximum empe a u e o con ol o his ac o . We
spa ially a e age Dayme pixels using municipali y bounda ies o es ima e o al p ecipi a ion le els
and maximum empe a u e o each municipali y’s su ace om 2012 o 2021. Appendix Figu e A.2
illus a es he Dayme esolu ion compa ed o one o he smalles municipali ies in Mexico.
The amoun o p ecipi a ion and empe a u e a ies h oughou he yea and ac oss he Mex-
ican e i o y (Figu es 2and A.3), suppo ing he ex e nal alidi y o ou analyses. The d y season
spans app oxima ely om No embe o Ap il and is cha ac e ized by lowe p ecipi a ion le els
ac oss mos o he coun y, wi h Janua y (le ) exempli ying his pe iod’s condi ions. Con e sely,
mos o he coun y’s ainy season ex ends om la e May o Oc obe , when ain all signi ican ly
inc eases. Figu e 2( igh ) shows ha mon hly o al p ecipi a ion can each 300 millili e s pe
mon h.
Figu e 2: P ecipi a ion da a
Sou ce: ORNL DAAC- Dayme Ve sion 4 Re ision 1; au ho s’ o mula ion.
3.3 Local cha ac e is ics
Municipali y bounda ies a e ob ained om he 2010 Geos a is ical F amewo k e sion 5.0 (INEGI,
2010)11. We use indi idual-le el occupa ion da a om he Mexican Census conduc ed in 2010
(INEGI,2010) o dis inguish be ween ag icul u al and non-ag icul u al municipali ies. F om his
Census, we de ine ag icul u al municipali ies as hose whe e he pe cen age o he popula ion wo k-
ing in ag icul u e is abo e he na ional median (39 pe cen ). The dis ibu ion o his a iable is
shown in Appendix Figu e A.4. The municipali y-le el popula ion used o calcula e iolence a es
is also ob ained om his sou ce.
11The 2010 map is di ided in o 2,546 municipali ies. Since some municipali ies we e c ea ed a e his amewo k,
all da abases we e adjus ed o main ain hese 2,456 municipali ies.
13
3.3.1 Municipali y-speci ic g owing seasons
To examine whe he po en ial income loss in ag icul u e is a sou ce o inc eased domes ic iolence,
we cons uc a municipali y-speci ic indica o o whe he he shock coincides wi h he g owing
season o he municipali y’s main c ops in e ms o o al e enue. To iden i y he main c op, we
employ da a om he yea ly s a is ical epo s published by he Ag icul u al and Fishe y In o ma-
ion Se ice SIAP (2024). This epo con ains municipali y-by-c op-by-cycle yields and e enue.
We iden i ied he municipali y’s main c op-cycle12 based on he agg ega e e enue be ween 2010
and 2018. Nex , we iden i y he ele an g owing season o each c op using he manuals o g ow-
ing each c op published by he Na ional Ins i u e o Fo es y, Ag icul u al and Li es ock Resea ch
(INIFAP). We complemen ed his in o ma ion wi h echnical documen s om he USDA and he
FAO. A de ailed desc ip ion o he da a sou ces employed o iden i y he g owing season o each
c op is displayed in Table A.1.
3.3.2 S a e capaci y
We cons uc a s a e capaci y index wi h municipal-le el a ia ion. Based on p e ious wo ks seeking
o cap u e subna ional di e ences in s a e capaci y, ou index is based on wo co e dimensions:
adminis a ion and ex ac ion (Soi e ,2012).
We employ sewe age and ap wa e co e age and local s a e employees pe capi a as p ox-
ies o adminis a ion capabili ies. Da a o hese a iables come om he Censo de Población y
Vi ienda 2010 epo ed by INEGI. Ex ac ion capaci y is p oxied wi h iscal au onomy, measu ing
municipali y axes collec ed as a sha e o o al esou ces. This in o ma ion is ob ained om Fi-
nanzas Públicas Es a ales y Municipales (EFIPEM). All a iables a e s anda dized and hen used
o c ea e an addi i e index. This index hen anges be ween -3.13 and 1.65, whe e lowe numbe s
indica e wo se s a e capaci y.
4 Empi ical amewo k
The i s s ep in ou empi ical s a egy is o quan i y he causal impac o d ough on IPV. We
ollow he s anda d clima e econome ics li e a u e (Dell e al.,2014;Hsiang,2016) and c ea e a
municipali y-le el d ough exposu e me ic ha coun s unexpec ed days wi hou ain wi hin speci ic
e e ence pe iods. We es ima e he ollowing equa ion:
DV m =
5
X
b=1
χbDb,m + Λm +γs( ) + δmy +θ +εm (1)
12Di e en egions migh g ow he same c op a di e en cycles due o ag o-clima ic di e ences. Fo ins ance, as
shown in Table A.1, he majo i y o coun ies’ mos impo an non-pe ennial c op is co n g ain, bu no all o hem
g ow be ween Ap il and Oc obe
14
In equa ion 1, he dependen a iable, DVm — he IPV hospi aliza ion, calls o police epo s
pe 100,000 inhabi an s in municipali y m, and ime (week o mon h) —is modeled as a lexible
semi-pa ame ic unc ion o d ough Dm . Speci ically, Db,m a e dummy a iables indica ing he
numbe o abno mal days wi hou p ecipi a ion in he pas mon h/ ou weeks, which allows o
po en ial non-linea i ies. Fo he mon hly a iables (police eco ds and ho line calls), Dm ep e-
sen s he days wi hou ain wi hin he cu en calenda mon h. Fo hospi al eco ds, which a e on
a weekly imescale and ma ched o weekly wea he indica o s, we use he sum o days wi hou ain
o e he pas ou weeks (weeks w∈[0,−3]), so Dm =P3
w=0 Daysno ainm, −w.13 We de ine hese
unexpec ed days based on de ia ions om his o ical p ecipi a ion no ms o each municipali y and
ime o he yea , inco po a ing municipali y-by-yea and da e- ixed e ec s. The me hodology is
designed o le e age exposu e o p ecipi a ion le els below hose o which egions ha e his o ically
adap ed. Λm includes con ols o o he ex eme wea he e en s in ime (days abo e/below he
5 h and 95 h pe cen ile o empe a u e and days abo e he 95 pe cen ile o p ecipi a ion). We
also include s a e-speci ic ime ends γs( ) o cap u e egional de elopmen pa e ns ha could
co ela e wi h p ecipi a ion ends. δmy a e municipali y-yea ixed e ec s and θ a e da e ixed
e ec s.
Municipali y ixed e ec s cap u e ime-in a ian cha ac e is ics speci ic o each municipali y,
such as local ins i u ions, socio-economic condi ions, and cul u al ac o s ha migh in luence IPV
a es. These e ec s accoun o any pe sis en di e ences ac oss municipali ies, including baseline
clima e and p e alence o d ough . In e ac ed wi h he yea , we allow mo e lexibili y o hese
di e en condi ions o change yea ly. Da e ixed e ec s (θ ) con ol o s anda d shocks ha a ec all
municipali ies simul aneously, such as na ional holidays, economic down u ns, o signi ican poli ical
e en s. By including hese, we adjus o empo al a ia ions ha could con ound he ela ionship
be ween d ough and IPV. Las ly, s a e-speci ic ime ends (γs( )) add ess egional pa e ns ha
e ol e, such as in as uc u e imp o emen s, economic de elopmen , demog aphic changes, and
e en egion-speci ic global wa ming, ensu ing ha ou esul s a e no d i en by long- e m ends
wi hin s a es ha could be co ela ed wi h bo h d ough and IPV. Ou iden i ica ion assump ion
is ha we a e cap u ing anomalous d ough days—–de ia ions om expec ed p ecipi a ion (i.e.,
p ecipi a ion ha could be p edic ed by he ixed e ec s desc ibed abo e)—–which a e he e o e
o hogonal o o he ele an d i e s o IPV.
5 Resul s
We begin ou analysis wi h he esul s om equa ion 1, which lexibly es ima es he ela ionship
be ween he in ensi y o anomalous d ough condi ions o e he pas mon h and he a e o IPV.
13Since hese wo measu es (days in he pas ou weeks s. days in he cu en mon h) ha e di e en uppe limi s
(28 s. 31), we use di e en bins o accu a ely cap u e hei a iabili y.
15
The dependen a iable in his analysis is ope a ionalized h ough h ee dis inc measu es, each
ep esen ing a di e en ype o in e ac ion be ween public se ices and IPV ic ims: hospi al
eco ds, police epo s, and ho line calls. As discussed abo e, o hospi al eco ds and ho line calls,
we can ob ain IPV a es, bu o police epo s, he da a agg ega es all o ms o domes ic iolence.
Only in he case o hospi al eco ds a e we able o sepa a e inciden s by he sex o he ic im.
5.1 Hospi al eco ds
Figu e 3shows he impac o anomalous d ough condi ions in he p eceding ou weeks on he
municipali y-le el a e o hospi al admissions o assaul cases in ol ing emale ic ims (a) and
male ic ims (b) in a gi en week. The y-axis ep esen s he o al anomalous days wi hou ain
in ha week and he h ee p e ious ones, g ouped in bins. The omi ed ca ego y includes weeks
wi h exposu e o less han a o al o 10 anomalous days o d ough du ing he p e ious weeks
−w, w ∈0,−3. The igu e shows a mono onic ela ionship be ween d ough and iolence agains
men, while nondomes ic iolence agains women pla eaus a e 19 days o d ough . The a e o
emale hospi aliza ions o non-domes ic iolence inc eases by 0.2 when he numbe o days o
d ough in he pas ou weeks exceeds 26. This amoun s o a 7.4% inc ease om he sample mean
o 2.5 weekly cases pe 100,000 inhabi an s. The male non-domes ic iolence a e displays a ise o
0.43 in he d ies bin, which amoun s o a 14% inc ease om a sample mean o 3.0 weekly cases
pe 100,000 inhabi an s. O e all, d ough impac s nondomes ic iolence agains men and women,
wi h a s onge inc ease o men bu signi ican ly a ec ing all he popula ion.
In ima e pa ne iolence ells a di e en s o y, displaying a ma kedly gende ed ela ion
o d ough . The numbe o ic ims hospi alized e e y week in he municipali y inc eases as he
in ensi y o he unexpec ed lack o ain expe ienced in he p eceding ou weeks inc eases (Figu e
4(a)). The e is an a e age inc ease in hospi aliza ions o emale ic ims o IPV, up o 9% o
he sample mean, which ep esen s 5.2 weekly cases pe 100,000 inhabi an s. On he o he hand,
he e is no s a is ically signi ican impac on male hospi aliza ions o IPV (Figu e 4(b)). We ind
no disce nible impac s on hospi aliza ions o child mal ea men , as shown in Appendix Figu e
A.5.
16

(a) Female ic ims (b) Male ic ims
Figu e 3: Anomalous days o d ough and non-domes ic iolence
No es: The dependen a iables a e weekly hospi al admissions wi hin he Mexican public heal h sys em om 2012
o 2019, ca ego ized acco ding o ICD-10 codes. The eg ession model adjus s o o he ex eme wea he e en s,
including days wi h empe a u es abo e o below he 5 hand 95 hpe cen iles and days exceeding he 95 hpe cen ile
o p ecipi a ion. The model inco po a es municipali y-yea and da e ixed e ec s, s a e-speci ic ime ends, and
con ols o municipali y popula ion sizes. E o ba s ep esen 95% con idence in e als.
(a) Female ic ims (b) Male ic ims
Figu e 4: Anomalous days o d ough and IPV
No es: The dependen a iables a e weekly hospi al admissions wi hin he Mexican public heal h sys em om 2012
o 2019, ca ego ized acco ding o ICD-10 codes. The eg ession model adjus s o o he ex eme wea he e en s,
including days wi h empe a u es abo e o below he 5 hand 95 hpe cen iles and days exceeding he 95 hpe cen ile
o p ecipi a ion. The model inco po a es municipali y-yea and da e ixed e ec s, s a e-speci ic ime ends, and
con ols o municipali y popula ion sizes. E o ba s ep esen 95% con idence in e als.
The di e en ial impac o d ough on non-domes ic iolence can be unde s ood h ough he
lens o economic and psychological heo ies. Economic dis ess induced by d ough may heigh en
household s ess and con lic due o inancial insecu i y, as discussed in he economic li e a u e on
17
esou ce sca ci y and iolence in Sec ion 2.2. Men may exhibi inc eased agg ession as a coping
mechanism o s ess and inancial p essu es, beha io consis en wi h he social s ess heo y,
which posi s ha en i onmen al s ess can exace ba e agg essi e beha io s (Jia,2014). This ype
o iolence may mani es agains bo h male and emale ic ims. The mo e ma ked inc ease o men
could be due o he ac ha hey spend mo e ime in public spaces, leading o mo e in e ac ions
ha can po en ially end in iolence. Finally, in a model o a ional c iminal beha io , he ela i e
e u ns o c iminal ac i i y inc ease when income om legal ac i i ies dec eases. I some o hese
addi ional c imes commi ed in ol e iolence as a means o coe cion, he e is an expec ed inc ease
in agg essions agains men and women, wi h men once again being disp opo iona ely impac ed as
hey occupy mo e public spaces.
5.2 Police iles and ho line calls
Hospi al eco ds o IPV only cap u e cases se e e enough o wa an an encoun e wi h he heal h
sys em and hus ep esen “ he ip o he icebe g” o he p oblem. Hence, in his Sec ion, we
employ wo addi ional da a sou ces o show ha ou esul s using hospi al eco ds a e obus o
o he measu es o IPV.
Al e na i e measu es ha e hei d awbacks, oo. Fi s , in Mexico, IPV is a ely epo ed o
he police. The impuni y a e o his c ime is 98.6% na ionally, and i su passes 90% in e e y
s a e (ENVIPE,2022). Second, as no ed be o e, om police epo s and ho line calls, we canno
dis inguish he sex o he ic im. The e is no eliable da a on he pe cen age o IPV ic ims o
wi nesses who call 911 du ing o a e an episode. Howe e , in p inciple, his da a migh su e
om ewe biases han he police epo s since i connec s women wi h mo e se ices and can be
made by a wi ness, such as a neighbo o ela i e.
The esul s o implemen ing equa ion 1using mon hly da a om he sou ces discussed abo e
a e shown in Figu e 5. Police iles and ho line calls exhibi consis en pa e ns in esponse o
he numbe o unexpec ed days wi hou ain in he cu en mon h. In he highes bin, indica ing
mo e han 28 days wi hou ain, police epo s and ho line calls inc ease by 18 iles pe 100,000
inhabi an s and 44 calls pe 100,00 inhabi an s pe mon h. These ma ked inc eases di e ge 26%
and 21%, espec i ely, om he sample mean.
18
(a) Police iles (b) Ho line calls
Figu e 5: Anomalous days o d ough and domes ic iolence
No es:The dependen a iables a e mon hly police iles (a) and IPV ho line calls (b), g ouped based on hei da e
o occu ence. Police eco ds co e 2015-2021, and ho line calls include 2016-2021. The eg ession model adjus s
o o he ex eme wea he e en s, including days wi h empe a u es abo e o below he 5 hand 95 hpe cen iles
and days exceeding he 95 hpe cen ile o p ecipi a ion. The model inco po a es municipali y-yea and da e ixed
e ec s, s a e-speci ic ime ends, and con ols o municipali y popula ion sizes. E o ba s ep esen 95% con idence
in e als.
Ho line calls ace a signi ican challenge due o insu icien geog aphical co e age, limi ing
he abili y o conduc ep esen a i e egional he e ogenei y analysis. As an eme ging da a sou ce,
only eigh ou o 32 s a es p o ided da a o e a su icien ly long pe iod (2016-2021) o enable a
ixed e ec s es ima ion ha cap u es he impac o ex eme e en s. Howe e , his da ase o e s he
ad an age o including epo s on a ious o ms o iolence, which, due o hei p e alence, signi -
ican ly a ec women’s e e yday li es. I e eals a la ge po ion o he p o e bial icebe g. Figu e
6examines he incidence o some o hese c imes. While epo s o ape and sexual ha assmen do
no appea o be in luenced by d ough condi ions, epo s o sexual abuse inc ease mono onically,
wi h coe icien s becoming s a is ically signi ican in mon hs wi h mo e han 25 days o d ough .
The ise in sexual abuse epo s is pa icula ly conce ning: acco ding o Mexican legisla ion, he
dis inc ion be ween sexual abuse and ape hinges on whe he he e was aginal, anal, o o al pen-
e a ion14, bu bo h ac s in ol e non-consensual sexual con ac . Needless o say, bo h c imes can
in lic signi ican auma on women.
14A icles 178 and 179 o he penal code.
19
(a) Rape (b) Sexual Abuse
(c) Sexual ha assmen
Figu e 6: Anomalous days o d ough and o he o ms o GBV
No es: The dependen a iables a e exp essed as a a e pe 100,000 inhabi an s. Hospi al eco ds co e 2012-2019,
police eco ds co e 2015-2021, and ho line calls include 2016-2021. The eg ession model adjus s o o he ex eme
wea he e en s, including days wi h empe a u es abo e o below he 5 hand 95 hpe cen iles and days exceeding he
95 hpe cen ile o p ecipi a ion. The model inco po a es municipali y-yea and da e- ixed e ec s and s a e-speci ic
ime ends. E o ba s ep esen 95% con idence in e als.
Al oge he , we ind a obus ela ionship be ween d ough and IPV. Analogously o he
empe a u e- iolence link iden i ied in Mexico (Baysan e al.,2019), he con empo a y and linea
esponse o c ime o d ough iden i ied in his pape sugges s ha psychological ac o s ha e po-
en ial explana o y powe .15. Howe e , as discussed in Sec ion 2.2, mechanisms d i ing iolence
agains women may di e om ypical in e /in ag oup iolence. In he ollowing sec ions, we p o-
ide sugges i e e idence ha po en ial ag icul u al income loss is a po en ial mechanism d i ing
15(Baysan e al.,2019) ind ha he esponse o suicide o empe a u e in Mexico s ongly ma ches he esponses o
in e pe sonal and in e g oup iolence o empe a u e: he esponse is linea , con empo a y, s anda d ac oss egions,
no media ed by obse able economic ac o s o access o a condi ional cash ans e p og am, and only ba ely a ec ed
by g owing season empe a u es. The au ho s do no examine he ole o d ough .
20
5.4 Robus ness checks
To alida e he consis ency o ou main indings, we employ a binned speci ica ion linking he
a e age p ecipi a ion in a gi en week and he incidence o ou dependen a iables o in e es . We
use he same ixed e ec s as in ou baseline speci ica ion and con ol o empe a u e.
Figu e 10 displays he esul s o non-domes ic iolence, wi h panels (a) and (b) showing he
e ec o anomalous d ough days on weekly hospi al admissions o assaul cases in ol ing emale
and male ic ims, espec i ely. The esul s indica e a posi i e ela ionship be ween he numbe o
anomalous d ough days and he numbe o hospi al admissions o assaul cases, wi h a somewha
s onge pa e n o male ic ims, consis en wi h ou main indings. Figu e 11 ocuses on IPV, wi h
panels (a) and (b) showing he ela ionship be ween weekly a e age p ecipi a ion and IPV hospi al
admissions o emale and male ic ims, espec i ely. The esul s show ha lowe p ecipi a ion
le els a e associa ed wi h highe IPV hospi al admissions o emale ic ims, while no signi ican
ela ionship is obse ed o male ic ims, as in ou main speci ica ion.
(a) Assaul a es - emale ic ims (b) Assaul a es - male ic ims
Figu e 10: Anomalous days o d ough and non-domes ic iolence
No es: The dependen a iables a e weekly hospi al admissions wi hin he Mexican public heal h sys em om 2012 o
2019, ca ego ized acco ding o ICD-10 codes. The eg ession model adjus s o a e age empe a u e. The model inco -
po a es municipali y-yea and da e ixed e ec s, s a e-speci ic ime ends, and con ols o municipali y popula ion
sizes. E o ba s ep esen 95% con idence in e als.
27

(a) IPV a es - emale ic ims (b) IPV a es - male ic ims
Figu e 11: Weekly a e age p ecipi a ion and IPV
No es: The dependen a iables a e weekly hospi al admissions wi hin he Mexican public heal h sys em om 2012 o
2019, ca ego ized acco ding o ICD-10 codes. The eg ession model adjus s o a e age empe a u e. The model inco -
po a es municipali y-yea and da e ixed e ec s, s a e-speci ic ime ends, and con ols o municipali y popula ion
sizes. E o ba s ep esen 95% con idence in e als.
6 Conclusion
We ind obus e idence o an inc ease in domes ic iolence in Mexico when households ace d ough
condi ions. Ou indings a e obus o u ilizing h ee da a sou ces: hospi al eco ds, police iles,
and 911 calls. Ou esul s sugges ha a lowe municipal s a e capaci y le el is ela ed o a g ea e
IPV d ough impac due o a weake de e ence e ec .
While accoun ing o 8% o na ional GDP, ag icul u e is an essen ial sou ce o li elihood o
millions o people ac oss he coun y. I is also a highly clima e-sensi i e indus y and a male-
domina ed sec o . Ou esul s indica e ha d ough -induced dis up ions in ensi y IPV mo e in
ag icul u ally in ensi e a eas and when i hi s du ing he g owing season. We hypo hesize ha
ad e se wea he e en s may challenge he con en ional ole o men as p ima y ea ne s, po en ially
leading o inc eased iolence as a esponse mechanism. O e all, ou e idence aligns mo e wi h
s ess- ela ed o ex ac i e iolence in esponse o d ough du ing he g owing season in ag icul u al
egions han wi h economic models o exp essi e iolence.
Mexico is a middle-income coun y, o he bes o ou knowledge, weal hie han e e y o he
coun y whe e he link be ween d ough and IPV has been s udied. This ac sugges s ha economic
g ow h i sel will no shield women om he exace ba ed clima e impac s hey ace. While he e
migh be a human igh s basis o policy in e en ion o has en his con e gence, in a wo ld wi h
limi ed esou ces, he e migh be mo e scope o go e nmen in e en ion in dimensions o gende
inequali y ha a e unlikely o diminish in he medium e m unde cu en economic g ow h ends.
28
Thus, ou indings ha e se e al policy implica ions wi h ex e nal alidi y and ele ance o a high
sha e o he global popula ion. Fi s o all, disas e esponse p og ams should conside he dispa a e
impac s ha women ace, inco po a ing IPV p e en ion and a en ion p o ocols in o eme gency
esponse sys ems, especially o d ough . Ou indings also unde sco e he need o s eng hen us
and use o women’s eme gency esou ces. Be e epo ing and use o helplines and o he esou ces
will gene a e be e da a ha can con ibu e o diagnosing ulne abili y and a ge in e en ions.
Finally, unlike o he o ms o iolence ha may be clima e-sensi i e, mo e policing is unlikely o
mi iga e he e ec s o wea he ex emes on iolence agains women since his ype o iolence
usually does no ake place in public spaces. In Mexico, mo e han 9 ou o 10 cases o IPV agains
women happen a home (ENDIREH,2021). Changing social oles ega ding he accep abili y o
domes ic iolence and he male ole as p ima y ea ne s could be a signi ican s ep in mi iga ing he
clima e-gende nexus.
29
Re e ences
Abiona, O. and Koppens eine , M. F. (2018). The impac o household shocks on domes ic
iolence: E idence om anzania. IZA discussion pape .
Adams, A. E.,Tolman, R. M.,Bybee, D.,Sulli an, C. M. and Kennedy, A. C. (2012). The
impac o in ima e pa ne iolence on low-income women’s economic well-being: The media ing
ole o job s abili y. Violence Agains Women,18 (12), 1345–1367.
Agüe o, J. M. (2021). Co id-19 and he ise o in ima e pa ne iolence. Wo ld de elopmen ,
137, 105217.
Aguila -Gomez, S.,A ceo-Gomez, E. and De la C uz Toledo, E. (2019). Inside he black
box o child penal ies: Unpaid wo k and household s uc u e. A ailable a SSRN 3497089.
Aize , A. (2010). The gende wage gap and domes ic iolence. Ame ican Economic Re iew,
100 (4), 1847–1859.
Ande be g, D.,Raine , H.,Wadswo h, J. and Wilson, T. (2016). Unemploymen and
domes ic iolence: Theo y and e idence. The Economic Jou nal,126 (597), 1947–1979.
Angelucci, M. (2008). Lo e on he ocks: Domes ic iolence and alcohol abuse in u al mexico.
The BE Jou nal o Economic Analysis & Policy,8(1).
—and Hea h, R. (2020). Women empowe men p og ams and in ima e pa ne iolence. In AEA
Pape s and P oceedings, Ame ican Economic Associa ion 2014 B oadway, Sui e 305, Nash ille,
TN 37203, ol. 110, pp. 610–614.
A ias, P.,Bellouin, N.,Coppola, E.,Jones, R.,K inne , G.,Ma o zke, J.,Naik,
V.,Palme , M.,Pla ne , G.-K.,Rogelj, J.,Rojas, M.,Sillmann, J.,S o el mo,
T.,Tho ne, P.,T ewin, B.,Achu a Rao, K.,Adhika y, B.,Allan, R.,A mou , K.,
Bala, G.,Ba imalala, R.,Be ge , S.,Canadell, J.,Cassou, C.,Che chi, A.,Collins,
W.,Collins, W.,Conno s, S.,Co i, S.,C uz, F.,Den ene , F.,De eczynski, C.,
Di Luca, A.,Diongue Niang, A.,Doblas-Reyes, F.,Dosio, A.,Dou ille, H.,Engel-
b ech , F.,Ey ing, V.,Fische , E.,Fo s e , P.,Fox-Kempe , B.,Fugles ed , J.,
Fy e, J.,Gille , N.,Gold a b, L.,Go ode skaya, I.,Gu ie ez, J.,Hamdi, R.,
Hawkins, E.,Hewi , H.,Hope, P.,Islam, A.,Jones, C.,Kau man, D.,Kopp, R.,
Kosaka, Y.,Kossin, J.,K ako ska, S.,Lee, J.-Y.,Li, J.,Mau i sen, T.,Maycock,
T.,Meinshausen, M.,Min, S.-K.,Mon ei o, P.,Ngo-Duc, T.,O o, F.,Pin o, I.,
Pi ani, A.,Ragha an, K.,Ranasinghe, R.,Ruane, A.,Ruiz, L.,Sallée, J.-B.,Sam-
se , B.,Sa hyend ana h, S.,Sene i a ne, S.,Sö ensson, A.,Szopa, S.,Takayabu,
I.,T éguie , A.-M., an den Hu k, B.,Vau a d, R., on Schuckmann, K.,Zaehle,
S.,Zhang, X. and Zick eld, K. (2021). Clima e Change 2021: The Physical Science Basis.
Con ibu ion o Wo king G oup I o he Six h Assessmen Repo o he In e go e nmen al Panel
on Clima e Change: Technical summa y, Camb idge, Uni ed Kingdom and New Yo k, NY, USA:
Camb idge Uni e si y P ess, pp. 33–144.
A o a-Jonsson, S. (2011). Vi ue and ulne abili y: Discou ses on women, gende and clima e
change. Global en i onmen al change,21 (2), 744–751.
Banxico (2022). Sequía en méxico y su po encial impac o en la ac i idad económica. Banco Cen al
México-In o me T imes al Ab il–Junio.
Baysan, C.,Bu ke, M.,González, F.,Hsiang, S. and Miguel, E. (2019). Non-economic
ac o s in iolence: E idence om o ganized c ime, suicides and clima e in mexico. Jou nal o
Economic Beha io & O ganiza ion,168, 434–452.
Bedna -F iedl, B.,Biesb oek, R.,Schmid , D.,Alexande , P.,Bø sheim, K.,Ca -
30
nice , J.,Geo gopoulou, E.,Haasnoo , M.,Cozanne , G. L.,Lionello, P.,Lipka,
O.,Möllmann, C.,Muccione, V.,Mus onen, T.,Piepenbu g, D. and Whi ma sh, L.
(2022). Clima e Change 2022: Impac s, Adap a ion and Vulne abili y. Con ibu ion o Wo king
G oup II o he Six h Assessmen Repo o he In e go e nmen al Panel on Clima e Change,
Camb idge, UK and New Yo k, USA: Camb idge Uni e si y P ess, pp. 1817–1927.
Bhalo a, S.,Kambhampa i, U.,Rawlings, S. and Siddique, Z. (2021). In ima e pa ne
iolence: The in luence o job oppo uni ies o men and women. The Wo ld Bank Economic
Re iew,35 (2), 461–479.
Bha acha ya, H. (2015). Spousal iolence and women’s employmen in india. Feminis Eco-
nomics,21 (2), 30–52.
Blakeslee, D.,Chau ey, R.,Fishman, R.,Malghan, D. and Malik, S. (2021). In he hea
o he momen : Economic and non-economic d i e s o he wea he -c ime ela ionship. Jou nal
o Economic Beha io & O ganiza ion,192, 832–856.
Bobonis, G. J.,González-B enes, M. and Cas o, R. (2013). Public ans e s and domes ic
iolence: The oles o p i a e in o ma ion and spousal con ol. Ame ican Economic Jou nal:
Economic Policy,5(1), 179–205.
Boes en, J. (2012). The s a e and iolence agains women in pe u: In e sec ing inequali ies and
pa ia chal ule. Social Poli ics,19 (3), 361–382.
Bosland, P. and Walke , S. (2014). G owing chiles in new mexico. h ps://pubs.nmsu.edu/
_h/H230/e.
Boye , A.,Meije , S. and Gilligan, M. (2020). Ad ancing gende in he en i onmen : Explo -
ing he iple nexus o gende inequali y, s a e agili y, and clima e ulne abili y. Washing on,
DC: IUCN & USAID.
Bu ke, M.,Hsiang, S. M. and Miguel, E. (2015). Clima e and con lic . Annu. Re . Econ.,
7(1), 577–617.
Campbell, J. C. (2002). Heal h consequences o in ima e pa ne iolence. The lance ,359 (9314),
1331–1336.
Ca uso, R.,Pe a ca, I. and Ricciu i, R. (2016). Clima e change, ice c ops, and iolence:
E idence om indonesia. Jou nal o Peace Resea ch,53 (1), 66–83.
Chowdhu y, J. R.,Pa ida, Y. and Goel, P. A. (2021). Does inequali y-adjus ed human
de elopmen educe he impac o na u al disas e s? a gende ed pe spec i e. Wo ld De elopmen ,
141, 105394.
CONAGUA (2022). Es adís icas del agua en méxico 2021. Ciudad de México: Sec e a ía del Medio
Ambien e y Recu sos Na u ales. Ob enido de h ps://sina. conagua. gob. mx/publicaciones.
Cook, B. I.,Mankin, J. S. and Anchukai is, K. J. (2018). Clima e change and d ough : F om
pas o u u e. Cu en Clima e Change Repo s,4, 164–179.
Cools, S.,Fla ø, M. and Ko sadam, A. (2020). Rain all shocks and in ima e pa ne iolence
in sub-saha an a ica. Jou nal o Peace Resea ch,57 (3), 377–390.
—and Ko sadam, A. (2017). Resou ces and in ima e pa ne iolence in sub-saha an a ica.
Wo ld De elopmen ,95, 211–230.
Coope , M.,Sandle , A.,Vi ellozzi, S.,Lee, Y.,Seymou , G.,Haile, B. and Azza i,
C. (2021). Re-examining he e ec s o d ough on in ima e-pa ne iolence. PLoS one,16 (7),
e0254346.
C oppens ed , A.,Golds ein, M. and Rosas, N. (2013). Gende and ag icul u e: Ine i-
ciencies, seg ega ion, and low p oduc i i y aps. The Wo ld Bank Resea ch Obse e ,28 (1),
31
79–109.
Cullen, C. (2023). Me hod ma e s: The unde epo ing o in ima e pa ne iolence. The Wo ld
Bank Economic Re iew,37 (1), 49–73.
Dai, A. (2013). Inc easing d ough unde global wa ming in obse a ions and models. Na u e
clima e change,3(1), 52–58.
Dell, M.,Jones, B. F. and Olken, B. A. (2014). Wha do we lea n om he wea he ? he new
clima e-economy li e a u e. Jou nal o Economic li e a u e,52 (3), 740–798.
Díaz, J.-J. and Salda iaga, V. (2023). A d op o lo e? ain all shocks and spousal abuse:
e idence om u al pe u. Jou nal o heal h economics,89, 102739.
Ellsbe g, M.,Heise, L.,Pena, R.,Agu o, S. and Wink is , A. (2001). Resea ching do-
mes ic iolence agains women: me hodological and e hical conside a ions. S udies in amily
planning,32 (1), 1–16.
ENDIREH (2021). Encues a Nacional sob e la Dinámica de las Relaciones en los Hoga es 2021
ENDIREH P incipales Resul ados. Inegi.
ENVIPE (2022). Encues a Nacional de Vic imización y Pe cepción sob e Segu idad Pública. Inegi.
Eps ein, A.,Benda id, E.,Nash, D.,Cha lebois, E. D. and Weise , S. D. (2020). D ough
and in ima e pa ne iolence owa ds women in 19 coun ies in sub-saha an a ica du ing 2011-
2018: A popula ion-based s udy. PLoS Medicine,17 (3), e1003064.
E en, B. and Keskin, P. (2018). Fo be e o o wo se?: Educa ion and he p e alence o
domes ic iolence in u key. Ame ican Economic Jou nal: Applied Economics,10 (1), 64–105.
—and —(2021). Female employmen and in ima e pa ne iolence: e idence om sy ian e ugee
in lows o u key. Jou nal o De elopmen Economics,150, 102607.
—and —(2024). T ade-o s? he impac o w o accession on in ima e pa ne iolence in cambodia.
Re iew o Economics and S a is ics, pp. 1–12.
Eswa an, M. (2018). S a e Complici y in he Sexual Assaul o Women: The Fa e o Cassand a?
Tech. ep., UBC Wo king Pape , Vancou e School o Economics, Uni e si y o B i ish ....
—and Malho a, N. (2011). Domes ic iolence and women’s au onomy in de eloping coun ies:
heo y and e idence. Canadian Jou nal o Economics/Re ue canadienne d’économique,44 (4),
1222–1263.
FAO (2019). México se p epa a pa a mi iga mejo las sequías. h ps://www. ao.o g/mexico/
no icias/de ail-e en s/es/c/1252317/.
—(2020). P eassu e in wa e esou ces. h ps://da a.apps. ao.o g/aquas a /?lang=en.
—(2021). The impac o disas e s and c ises on ag icul u e and ood secu i y: 2021. I aly: Food
and Ag icul u e O ganiza ion o he Uni ed Na ions Rome, I aly.
—(2022). Giews coun y b ie mexico. h ps://www. ao.o g/giews/coun yb ie /coun y/
MEX/pd _a chi e/MEX_A chi e.pd .
Fa ema, S. R.,Islam, M. S.,Eas , L. and Ushe , K. (2019). Women’s heal h- ela ed ulne -
abili ies in na u al disas e s: a sys ema ic e iew p o ocol. BMJ open,9(12), e032079.
Ge e s, A.,Musuya, T. and Bukuluki, P. (2020). Why clima e change uels iolence agains
women: Uni ed na ions de elopmen p og amme.
Heilmann, K.,Kahn, M. E. and Tang, C. K. (2021). The u ban c ime and hea g adien in
high and low po e y a eas. Jou nal o Public Economics,197, 104408.
Hewage, S.,Banda a, M.,Ra hnayake, D.,Resea ch, H. K. A. and Ins i u e, T. (2022).
Local Hyb id Chili Cul i a ion and D y Chili P oduc ion: Towa ds Spicing Up he P ocess. Oc-
casional publica ion, Hec o Kobbekaduwa Ag a ian Resea ch and T aining Ins i u e.
32

Hi onen, K.,Gilligan, D. O.,Leigh , J.,Tambe , H. and Villa, V. (2023). Do ul a-poo
g adua ion p og ams build esilience agains d ough s? E idence om u al E hiopia. In l Food
Policy Res Ins .
Hsiang, S. (2016). Clima e econome ics. Annual Re iew o Resou ce Economics,8, 43–75.
Hsiang, S. M.,Bu ke, M. and Miguel, E. (2013). Quan i ying he in luence o clima e on
human con lic . Science,341 (6151), 1235367.
ICAMEX (2023a). Cul i o de ji oma e. h ps://icamex.edomex.gob.mx/ji oma e.
—(2023b). Cul i o de papa. h ps://icamex.edomex.gob.mx/papa#:~: ex =La%20siemb a%
20se%20debe%20de,oc ub e%20al%2015%20de%20diciemb e.
ILO (2012). El empleo de las muje es u ales en amé ica la ina. In e na ional Labou O ganiza ion.
IMCO (2024). La ges iÓn del agua no es una p io idad en mÉxico. h ps://imco.o g.mx/
la-ges ion-del-agua-no-es-una-p io idad-en-mexico/.
INEGI (2010). Censo de población y i ienda 2010. Ins i u o Nacional de Es adís ica y Geog a ía.
INIFAP (2012). Toma e. h p://www.ini apci ne.gob.mx/Biblio eca/Paque es2012/149.
pd .
—(2013a). F ijol. h p://www.ini apci ne.gob.mx/Biblio eca/Paque es2012/137.pd .
—(2013b). TecnologÍa pa a la p oducciÓn de a ena o aje a en el al iplano de amaulipas. h p:
//www.ini apci ne.gob.mx/Biblio eca/Publicaciones/956.pd .
—(2015a). Cul i e pas o ye g ass pa a la alimen aciÓn del ganado en la Época in e nal en el
no e y cen o de amaulipas. h p://www.ini apci ne.gob.mx/E en os/2015/Siemb a+Rye+
G ass.pd .
—(2015b). Paque e ecnológico pa a maíz o aje o. h ps://www.p oducechihuahua.o g/paqs/
PT-0007Maiz2.pd .
—(2024a). Biblio eca digi al del ins i u o nacional de in es igaciones o es ales, ag ícolas y pecua -
ias. h ps:// un.ini ap.gob.mx/Biblio ecaWeb/_Con en .
—(2024b). Calabaci a. h p://oiapes.saga hpa.sono a.gob.mx/paq- ec/paq-calabaci a.
pd .
—(2024c). Sandía. h p://oiapes.saga hpa.sono a.gob.mx/paq- ec/paq-sandia.pd .
—(2024d). So go o aje o. h p://oiapes.saga hpa.sono a.gob.mx/paq- ec/
paq-so go- o aje o.pd .
Jia, R. (2014). Wea he shocks, swee po a oes and peasan e ol s in his o ical china. The Eco-
nomic Jou nal,124 (575), 92–118.
Kee hi a ne, S. and Tol, R. S. (2018). Impac o na u al disas e s on income inequali y in s i
lanka. Wo ld De elopmen ,105, 217–230.
Kuzma, S.,Saccoccia, L. and Che ock, M. (2023). 25 coun ies, housing one-qua e o he
popula ion, ace ex emely high wa e s ess. Wo ld Resou ces Ins i u e. h ps://www. w i. o g.
Leslie, E. and Wilson, R. (2020). Shel e ing in place and domes ic iolence: E idence om calls
o se ice du ing co id-19. Jou nal o public economics,189, 104241.
Logie, C. H.,Daniel, C.,Ahmed, U. and Lash, R. (2017). ‘li e unde he en is no sa e,
especially o young women’: unde s anding in e sec ional iolence among in e nally displaced
you h in leogane, hai i. Global heal h ac ion,10 (sup2), 1270816.
Miedzian, M. (2002). Boys will be boys: B eaking he link be ween masculini y and iolence.
Lan e n Books.
Mille , A. R.,Segal, C. and Spence , M. K. (2024). E ec s o he co id-19 pandemic on
domes ic iolence in los angeles. Economica,91 (361), 163–187.
33
Munala, L.,Allen, E. M.,F ede ick, A. J. and Ng˜
unji i, A. (2023a). Clima e change,
ex eme wea he , and in ima e pa ne iolence in eas a ican ag a ian-based economies. In e -
na ional jou nal o en i onmen al esea ch and public heal h,20 (23), 7124.
—,—,—and Ng˜
unji i, A. (2023b). Clima e change, ex eme wea he , and in ima e pa ne
iolence in eas a ican Ag a ian-Based economies. In J En i on Res Public Heal h,20 (23).
Neumaye , E. and Plümpe , T. (2007). The gende ed na u e o na u al disas e s: The impac
o ca as ophic e en s on he gende gap in li e expec ancy, 1981–2002. Annals o he associa ion
o Ame ican Geog aphe s,97 (3), 551–566.
Nguyen, H. T. (2019). Gende ed ulne abili ies in imes o na u al disas e s: male- o- emale
iolence in he philippines in he a e ma h o supe yphoon haiyan. Violence Agains Women,
25 (4), 421–440.
Odu o, A. D.,Dee e, C. D. and Ca anza i e, Z. B. (2015). Women’s weal h and in ima e
pa ne iolence: Insigh s om ecuado and ghana. Feminis Economics,21 (2), 1–29.
ONU (). Jo nale as ag ícolas mexicanas luchan po sus de echos y la p o ección del medio ambien e.
ONU Muje es. Amé ica La ina y El Ca ibe.
Pa ma , P. K.,Jin, R. O.,Walsh, M. and Sco , J. (2019). Mo ali y in ohingya e ugee
camps in bangladesh: his o ical, social, and poli ical con ex . Sexual and ep oduc i e heal h
ma e s,27 (2), 39–49.
Quisumbing, A. R. and Doss, C. R. (2021). Gende in ag icul u e and ood sys ems. Handbook
o ag icul u al economics,5, 4481–4549.
Regis o Ag a io Nacional (2019). Es adís ica con pe spec i a de géne o, abulados.
h p://www. an.gob.mx/ an/index.php/sis emas-de-consul a/es adis ica-ag a ia/
es adis ica-con-pe spec i a-de-gene o.
Rincón-A alos, P.,Khouakhi, A.,Mendoza-Cano, O.,C uz, J. L.-D. l. and Pa edes-
Bonilla, K. M. (2022). E alua ion o sa elli e p ecipi a ion p oduc s o e mexico using google
ea h engine. Jou nal o Hyd oin o ma ics,24 (4), 711–729.
Roncoli, C.,Ing am, K. and Ki shen, P. (2001). The cos s and isks o coping wi h d ough :
li elihood impac s and a me s1 esponses in bu kina aso. Clima e esea ch,19 (2), 119–132.
Sabia, J. J.,Dills, A. K. and DeSimone, J. (2013). Sexual iolence agains women and labo
ma ke ou comes. Ame ican Economic Re iew,103 (3), 274–278.
Sekh i, S. and S o eyga d, A. (2014). Dow y dea hs: Response o wea he a iabili y in india.
Jou nal o de elopmen economics,111, 212–223.
SIAP (2024). Anua io es adís ico de la p oducción ag ícola. Se icio de In o mación Ag oalimen-
a ia y Pesque a.
Soi e , H. D. (2012). Measu ing s a e capaci y in con empo a y la in ame ica. Re is a de Ciencia
Polí ica,32 (3), 585–598.
S&P (2023). Más es ados mexicanos pod ían e se a ec ados po es és hí-
d ico en 2050. h ps://www.spglobal.com/ a ings/es/pd -a icles/2023/
2023-04-04-mas-es ados-mexicanos-pod ian- e se-a ec ados-po -es es-hid ico-en-2050.
S ia schi, M. M. and T ako, I. (2024). Gende iolence, en o cemen , and human capi al:
E idence om women’s jus ice cen e s in pe u. Jou nal o De elopmen Economics, p. 103262.
Te azas-Ca illo, E. C. and McWhi e , P. T. (2015). Employmen s a us and in ima e
pa ne iolence among mexican women. Jou nal o in e pe sonal iolence,30 (7), 1128–1152.
Tho n on, M.,Sh es ha, R.,Wei, Y.,Tho n on, P.,Kao, S.-C. and Wilson, B. (2022).
Dayme : Daily su ace wea he da a on a 1-km g id o no h ame ica, e sion 4 1.
34
Thu s on, A. M.,S öckl, H. and Rangana han, M. (2021). Na u al haza ds, disas e s and
iolence agains women and gi ls: a global mixed-me hods sys ema ic e iew. BMJ global heal h,
6(4), e004377.
Ud y, C. (1996). Gende , ag icul u al p oduc ion, and he heo y o he household. Jou nal o
poli ical Economy,104 (5), 1010–1046.
UN Women (2022). Gende , clima e and con lic analysis in somalia and assessmen o oppo -
uni ies o clima e sma ag icul u e and li elihood oppo uni ies o c isis-a ec ed and a - isk
women in somalia.
USDA (2024). C op calenda s o mexico, cen al ame ica and he ca ibbean. Uni ed S a es De-
pa men o Ag icul u e,h ps://ipad. as.usda.go / ssiws/al/c op_calenda /ca.aspx.
Van Aels , K. and Hol oe , N. (2016). In e sec ions o gende and ma i al s a us in accessing
clima e change adap a ion: E idence om u al anzania. Wo ld De elopmen ,79, 40–50.
an Daalen, K. R.,Kallesøe, S. S.,Da ey, F.,Dada, S.,Jung, L.,Singh, L.,Issa, R.,
Emilian, C. A.,Kuhn, I.,Keygnae , I. e al. (2022). Ex eme e en s and gende -based
iolence: a mixed-me hods sys ema ic e iew. The Lance Plane a y Heal h,6(6), e504–e523.
Wa , P. and Aung, L. L. (2019). Po e y and inequali y impac o a na u al disas e : Myanma ’s
2008 cyclone na gis. Wo ld De elopmen ,122, 446–461.
Wo ld Bank (2022a). Ag icul u e, o es y, and ishing, alue added (% o gdp). Accessed: 2024-
05-19.
Wo ld Bank (2022b). Violence agains women and gi ls – wha he da a ell us. h ps:
//gende da a.wo ldbank.o g/en/da a-s o ies/o e iew-o -gende -based- iolence.
Wo ld Bank (2023). Wa e . h ps://www.bancomundial.o g/es/ opic/wa e /o e iew.
Ylipaa, J.,Gab ielsson, S. and Je neck, A. (2019). Clima e change adap a ion and gende
inequali y: insigh s om u al ie nam. Sus ainabili y,11 (10), 2805.
Za e i, E. D.,Damania, R. and Engle, N. (2023). D ough s and de ici s. Wo ld Bank, Wash-
ing on, DC.
35
A Online Appendix
Figu e A.1: Le el o economic de elopmen o coun ies wi h quan i a i e s udies o he causal
impac o d ough s on IPV
No es: The published e e enced wo ks a e include Sekh i and S o eyga d (2014) (India), Cools and Ko sadam
(2017) (Sub-Saha an A ica) and Díaz and Salda iaga (2023) (Pe u). Wo king pape s e e enced include Abiona and
Koppens eine (2018) (Tanzania) and Hi onen e al. (2023) (E hiopia).
36