Chia ella, C is ina e al.
A icle — Published Ve sion
Impac s o la ge-scale o es y in es men s on
neighbo ing small-scale ag icul u e in no he n
Mozambique
Land Use Policy
P o ided in Coope a ion wi h:
Leibniz Ins i u e o Ag icul u al De elopmen in T ansi ion Economies (IAMO), Halle (Saale)
Sugges ed Ci a ion: Chia ella, C is ina e al. (2024) : Impac s o la ge-scale o es y in es men s on
neighbo ing small-scale ag icul u e in no he n Mozambique, Land Use Policy, ISSN 1873-5754,
Else ie , Ams e dam, Vol. 145, pp. 1-15,
h ps://doi.o g/10.1016/j.landusepol.2024.107251 ,
h ps://www.sciencedi ec .com/science/a icle/pii/S0264837724002047
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/302542
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 p://c ea i ecommons.o g/licenses/by/4.0/
Land Use Policy 145 (2024) 107251
A ailable online 29 July 2024
0264-8377/© 2024 The Au ho (s). Published by Else ie L d. This is an open access a icle unde he CC BY license (h p://c ea i ecommons.o g/licenses/by/4.0/).
Impac s o la ge-scale o es y in es men s on neighbo ing small-scale
ag icul u e in no he n Mozambique
C is ina Chia ella
a
,
g
,
*
, Philippe Ru in
a
,
b
, Dilini Abeygunawa dane
c
, Adia Bey
a
,
S´
a Noguei a Lisboa
d
,
e
, Helde Za ale
d
, Pa ick Mey oid
a
,
a
Ea h and Li e Ins i u e, UCLou ain, Lou ain-la-Neu e 1348, Belgium
b
Geog aphy Depa men , Humbold -Uni e si ¨
a zu Be lin, Be lin 10117, Ge many
c
Leibniz Ins i u e o Ag icul u al De elopmen in T ansi ion Economies (IAMO), Halle (Saale) 06120, Ge many
d
Uni e sidade Edua do Mondlane, Mapu o, Mozambique
e
N’Lab, Ni idae, Mapu o, Mozambique
Fonds de la Reche che Scien i ique F.R.S.-FNRS, B ussels 1000, Belgium
g
In e na ional Fund o Ag icul u al De elopmen (IFAD), Rome, I aly
ARTICLE INFO
Keywo ds:
Fo es y plan a ions
La ge-scale land acquisi ions
Ag icul u al employmen
Remo e sensing
Ag icul u al yields
ABSTRACT
Fo es y plan a ions can po en ially os e u al de elopmen and mi iga e en i onmen al h ea s, bu hei
impac s on neighbo ing peoples’li elihood s a egies a e ambiguous. Fo es y plan a ions a e pa icula ly
impo an in Mozambique, whe e a na ional s a egy aims o es ablish one million hec a es o o es s by 2030,
ocusing on Miombo eco egions in he p o inces o Niassa, Cabo Delgado, Nampula and Zambezia. This pape
e alua es he causal e ec s o la ge-scale o es y in es men s in no he n Mozambique on smallholde s’ a m
size, c op p oduc i i y, and employmen . We ake ad an age o a emo e sensing app oach ha p oduced maps o
o es y plan a ions and hei expansion ajec o ies om 2001 o 2017 and combine hem wi h da a om wo
geo e e enced na ionally- ep esen a i e ag icul u al su eys adminis e ed in 2007 and 2017. Using a di e ence-
in-di e ence app oach, we e alua e he e ec s o exposu e o o es y plan a ions es ablished a e 2007, de ined
by he p esence o newly es ablished and expansion o exis ing plan a ions and hei dis ance o households
wi hin a 20-km bu e . We ind ha households exposed o o es y plan a ions inc eased hei plan ed a eas bu
did no change hi ed a m employmen , which was accompanied by a dec ease in c op yields. The heads o
households close o o es y plan a ions we e also less likely o wo k in ag icul u e as hei main ac i i y,
especially as sala y wo ke s, and mo e likely o be sel -employed and employed in he non a m sec o . This s udy
con ibu es o an imp o ed unde s anding o local dynamics esul ing om o es y in es men s, which ha e
c i ical implica ions o in es men a ge ing and sus ainable land use planning.
1. In oduc ion
Global and na ional s a egies aiming o coun e g eenhouse gas
emissions, deg ada ion o na u al o es s, and he loss o biodi e si y,
inc easingly ely on ee plan a ions as a means o achie e hese a ge s.
These s a egies o en couple ecosys em se ices wi h he p o ision o
jobs and income sou ces o he local popula ion. This p omp s go -
e nmen s o alloca e land o o es y companies, al hough such alloca-
ions a e some imes in con lic wi h he in e es s o communi ies and
hei land igh s (Boone, 2012; Bleye e al., 2016; Kalabamu, 2019;
Rasmussen and Lund, 2018). In Mozambique, a Na ional Re o es a ion
S a egy se in 2009 aimed a inc easing comme cial o es plan a ion
a ea o 1 million hec a es (ha) in 2030, p ima ily h ough la ge-scale
co po a e plan a ions, ocusing on Miombo eco egions in he p o -
inces o Niassa, Cabo Delgado, Nampula and Zambezia ( e e ed o
he ea e as no he n Mozambique
1
). The ambi ious manda e included
he c ea ion o 250,000 pe manen jobs (Wo ld Bank, 2016). Howe e ,
he ac ual impac o o es y plan a ions on local popula ions’wel a e,
especially he spillo e s on ag icul u e and employmen , emains an
open empi ical ques ion which his esea ch aims o answe .
* Co esponding au ho a : Ea h and Li e Ins i u e, UCLou ain, Lou ain-la-Neu e 1348, Belgium.
E-mail add ess: [email p o ec ed] (C. Chia ella).
1
Adminis a i ely, Zambezia p o ince belongs o cen al Mozambique, while Niassa, Cabo Delgado and Nampula belong o no he n Mozambique. Howe e , we
classi y Zambezia as no he n Mozambique because Zambezia’s ag oecological condi ions a e mo e like he no he n egion han he cen al egion.
Con en s lis s a ailable a ScienceDi ec
Land Use Policy
jou nal homepage: www.else ie .com/loca e/landusepol
h ps://doi.o g/10.1016/j.landusepol.2024.107251
Recei ed 6 Decembe 2023; Recei ed in e ised o m 19 Ap il 2024; Accep ed 21 June 2024
Land Use Policy 145 (2024) 107251
2
The e ec s o la ge-scale land acquisi ions (LSLA) on neighbo ing
small-scale ag icul u e and local communi ies’li elihoods is a opic o
cons an deba e in he li e a u e. S udies ha e explo ed hei e ec s on a
a ie y o ou comes including displacemen o smallholde s o o he
a eas o economic sec o s, land ma ke s, land use, ma ke access, labo
abso p ion, and small-scale a me s’ag icul u al p oduc i i y (Mal-
kam¨
aki e al., 2018). Bu he di ec ion and he signi icance o he e ec s
emain a om consensual. Pa o he empi ical e idence sugges s ha
la ge-scale in es men s may ha e posi i e spillo e s in hei icini y,
such as inc eased ma ke oppo uni ies o high- alue c ops, inc eased
connec i i y, and lowe inpu cos s (Bu ke e al., 2020; Si ko e al.,
2018), inc eased cul i a ed a ea and yields (Lay e al., 2021), inc eased
incomes (He mann, 2017), inc eased employmen c ea ion in he su -
oundings (Deininge and Xia, 2016), and po e y educ ion (A onso
and Mille , 2021; Phimma ong and Keenan, 2020). O he s udies
documen ha he p oximi y and exposu e o la ge-scale in es men s
does no con ibu e o employmen gene a ion and p o ides mode a e
bene i s o small-scale pa cels in he icini y (An i, 2021; Ali e al., 2019;
Jung and Hajja , 2023), does no lead o inc eased access o ma ke s,
inc eased cul i a ed a eas o inc eased ag icul u al p o i s (Deininge
and Xia, 2016), displaces smalle -scale a me s (Nol e and Os e meie ,
2017; Zaeh inge e al., 2021) and would lead o inc eased wel a e
inequali y (Phimma ong and Keenan, 2020). The e idence also sugges s
ha la ge-scale in es men s p omo e a ansi ion o c ops high in calo-
ies, bu low in nu i ional con en , o ien ed owa ds expo ma ke s,
which may displace he p oduc ion o adi ional local c ops, and lowe
g adually he die a y di e si y (Mülle e al., 2021).
These e ec s migh a y because o a se ies o ac o s, linked o he
con ex in which he in es men s ake place, as well as o he cha ac-
e is ics o he households and o he in es men s hemsel es, which
co e a wide a ie y o ac o s, business models, and land uses (Abey-
gunawa dane e al., 2022; Obe lack e al., 2021). Gi en his, we ocus on
one speci ic ype o la ge-scale in es men he e, which is he majo one
in e ms o land a ea occupied in no he n Mozambique, i.e. la ge-scale
o es y plan a ions ha ocus on wood p oduc ion (Bey and Mey oid ,
2021).
We aim o con ibu e o he LSLA knowledge base by e alua ing he
speci ic e ec s o la ge-scale o es y in es men s and hei expansion on
he wel a e o small-scale a me s in he su oundings. We ocus on
e alua ing hei e ec s on a mland expansion, c opland p oduc i i y
and labo . Fo each o hese ou comes, di e en mechanisms may lead o
opposi e e ec s. We discuss concep ually such possible mechanisms and
assess empi ically he ne di ec ion o hese changes.
The e alua ion o he impac s o la ge-scale in es men s on neigh-
bo ing landscapes and peoples’li elihoods is ypically challenged by
da a cons ain s. Se e al s udies ake ad an age o he la ge in es men s
egis e ed in he Land Ma ix da abase on land deals (Mülle e al., 2021;
Lay and Nol e, 2018), which p o ide in o ma ion on he main deals, bu
ha e limi ed da a on ac ual land uses o smalle deals ha a e imple-
men ed on he g ound. O he s udies ob ain he in o ma ion on land
acquisi ions om household su eys which may su e om an
unde - ep esen a ion o la ge-scale landholdings and a e also con-
s ained by sho ime pe iods in be ween su ey ounds (Deininge and
Xia, 2016).
We combine emo e sensing and household su ey da a o e alua e
he e ec s o o es y expansion on he wel a e o a me s loca ed in he
icini y. We use land use ajec o ies o ee plan a ion expansion in o
p io na u al ege a ion and c opland om 2001 o 2017 o no he n
Mozambique, ob ained h ough emo e sensing echniques ha dis in-
guish ee plan a ions om na u al ege a ion (Bey and Mey oid ,
2021). We combine his da a wi h wo geo e e enced
na ionally- ep esen a i e ag icul u al su eys o 2007 and 2017, ha
collec de ailed pa cel-le el in o ma ion on c op ypes, land manage-
men , p oduc ion and labo , among o he in o ma ion such as de-
mog aphic cha ac e is ics, asse owne ship, ood secu i y. Th ough a
di e ence-in-di e ence app oach, we e alua e he e ec s o all
o es y in es men s es ablished in he a ea on ou comes o ag icul u al
p oduc i i y and labo .
This s udy con ibu es me hodologically o he deba e on he impac s
o LSLA. We add ess common challenges in exis ing s udies, such as he
sho ime pe iods o e alua ions and he ep esen a i i y o he LSLA.
We do so by using a census o o es y plan a ions in no he n
Mozambique building on emo ely sensed da a p oduc s, which allow us
o obse e he expansion o he o es y plan a ions be ween 2007 and
2017, a a e oppo uni y in hese kinds o s udies. This in o ma ion also
con as s wi h p e ious s udies in ha he “ ea men ”o exposu e in-
o ma ion is he ac ual land use change, no he p esence o speci ic
deals o known companies o in es men s o ce ain cha ac e is ics. We
also con ibu e o he exis ing li e a u e o causal in e ence s udies ha
do no dis inguish LSLA by land use (Mülle e al., 2021; Deininge and
Xia, 2016), by disen angling he e ec s o LSLA o a speci ic ype o
in es men s, o es y plan a ions, and o p io land uses. The indings o
his esea ch con ibu e o a be e unde s anding o local dynamics o
o es y LSLA in no he n Mozambique, which has c i ical implica ions
o mo e inclusi e and sus ainable planning and de elopmen in he
a ea.
2. Land enu e dynamics in Mozambique’s o es y sec o
The i s o eign-owned plan a ions in Mozambique da e back o he
colonial pe iod in he ea ly o mid-20 h cen u y, p ima ily o com-
me cial pu poses by Po uguese colonize s. Since hen, successi e wa es
o in es o s a emp ed o es ablish plan a ions bu ailed and le , o
emained bu wi hou being success ul (K onenbu g Ga cía e al., 2022).
In ecen imes, p omo ing la ge-scale in es men s has become one o
he ag icul u al de elopmen models pu sued by he Go e nmen o
Mozambique (GoM) (No a and Ros´
a io, 2022). Be ween 2005 and 2008
o eign in es men companies pionee ing a new wa e we e se up. By
2009, Mozambique had 60,000 ha o la ge-scale comme cial plan ed
o es , di ec ly p o iding 3000 jobs (Se zedelo de Almeida and Delgado,
2019). In 2012, companies in Niassa only had been issued six Land Use
Righ s ce i ica es - he ea e e e ed o as DUATs om i s Po uguese
ac onym o Di ei o de Uso e Ap o ei amen o de Te a - co e ing abou
155,000 ha and in es ed abou USD 70 million (Wo ld Bank, 2016).
These a e mos ly monocul u e ee plan a ions o pine and eucalyp us.
The wo majo plan a ion companies wi h he g ea es in es men ha
ope a e in no he n and cen al Mozambique a e Po ucel Mozambique
and G een Resou ces (O lowski, 2016). Po ucel Mozambique ecei ed
land igh s co e ing 356,000 ha (Se zedelo de Almeida and Delgado,
2019).
The expansion o hese la ge-scale plan a ions has in e sec ed wi h
exis ing land enu e s uc u es and local communi y dynamics. Po ucel
Mozambique, o ins ance, adop ed a “mosaic”model, whe e wo hi ds
o he o al a ea a e plan ed and one hi d is ese ed o communi y use.
Wi h his app oach, communi y a ms end up su ounded by he plan-
a ion, po en ially wi h nega i e impac s on hese communi ies because
o he eucalyp us allelopa hic e ec .
2
P elimina y anecdo al e idence
sugges s his may be p omp ing a ming households o mo e o mo e
dis an a eas in sea ch o e ile land and wa e esou ces (O lowski,
2016), leading o po en ial con lic s and dis up ions in adi ional
ag icul u al p ac ices.
Cen al o hese dynamics a e land enu e a angemen s, which a e
he mos common sou ce o con lic s be ween o es companies and local
communi ies (Nhan umbo e al., 2013). All land in Mozambique is
publicly owned. The land i sel canno be sold, bu he GoM can g an
concessions o land use igh s h ough DUATs. The Land Law o 1997
2
Eucalyp us ees elease chemical compounds ha can in luence he g ow h
and de elopmen o o he plan s in he icini y. The chemicals eleased by
eucalyp us ees in o he en i onmen can ha e ei he inhibi o y o s imula o y
e ec s on he g ow h o neighbo ing plan s (Zhang and Fu, 2009).
C. Chia ella e al.
Land Use Policy 145 (2024) 107251
3
es ablished ha DUATs can be acqui ed ei he by indi idual pe sons o
local communi ies h ough cus oma y no ms, by good- ai h occupa ion
o a leas en yea s by na ional indi idual pe sons; o h ough he
au ho iza ion o an applica ion submi ed o Public Adminis a ion by
an indi idual o co po a e pe son (MozLegal, 2004). The Land Law
ecognized adi ional igh s o sys ems o cus oma y occupancy and he
ole o communi ies in managemen o na u al esou ces, esolu ion o
con lic s, he p ocess o i ling, and he de ini ion o bounda ies ha he
communi ies occupy.
Land use igh s o o eign companies, on he o he hand, ha equi e
DUATs o economic pu poses, a e subjec o he app o al o an in-
es men o land use plan. Companies a e g an ed i s a p o isional
DUAT, which subjec o he comple ion o he land use plan, a e g an ed
a de ini i e DUAT wi h a alidi y o up o 50 yea s, ha can be enewed
o u he 50 yea s. The igh s o land use may be ans e ed by in-
he i ance o by a public no a ial deed; and in he case o companies,
pending on au ho iza ion by he same en i y ha app o ed he DUAT
and on he ul ilmen o he land use plan.
DUATs a e issued o local communi ies o indi idual pe sons who
eques indi idual land use igh s a e he plo has been pa i ioned
om i s espec i e communi y land. The absence o a DUAT owne ship,
howe e , does no p e en land use igh s. The applica ion o a DUAT
o economic ac i i ies mus include a s a emen om local au ho i ies
ha con i ms ha he a ea is ee and has no occupan s a e consul a-
ion wi h he communi y.
Land mapped o o es ac i i ies by he GoM is o en al eady in use
by local communi ies, so conceding land use igh s equi es long con-
sul a ions and nego ia ions (Wo ld Bank, 2016). Ye , mul iple i egu-
la i ies ha e been p e iously epo ed wi h such consul a ions, such as
ailu e o unde ake he consul a ions, eco ds alsi ica ion, poo
consul a ion p ocesses, consul a ion wi h only one o se e al a ec ed
pa ies, co up ion ac s and b ibes, among o he s (Si oe e al., 2012;
Hanlon, 2002; Ve meulen and Co ula, 2010). E idence also sugges s
ha in Mozambique non-ce i ied
3
p i a e plan a ions a e less likely o
in ol e local communi ies in he consul a ion p ocesses (Degne e al.,
2022).
3. Concep ual amewo k
Figu e 1 p esen s h ee possible pa hways o he impac s o o es y
plan a ions on households’ou comes o in e es , which a e cul i a ed
a ea, c op ypes, yields, hi ed a m employmen , and household head’s
employmen sec o . On he i s pa hway, we specula e ha o es y
plan a ions may a ac inpu supplie s and make inpu s mo e accessible
and a o dable and open up ma ke oppo uni ies such as a ac ing
ade s. I such a pa hway was o domina e, households may be able o
a o d inpu s such as imp o ed seeds and mo e likely o plan high- alue
c ops. This could likely lead o inc eases in ou pu alue and highe
yields. I a ebound e ec occu s, his pa hway may also lead o an
expansion o cul i a ed a ea (Mey oid e al., 2018). I economic e u ns
a e g ea e , households may also be mo e likely o hi e mo e a m
wo ke s, and he household head may s ay in ag icul u e as an
own-accoun wo ke .
On a second pa hway ep esen ed in Figu e 1, o es y plan a ions
may di ec ly employ plan a ion wo ke s, so ha a posi i e e ec on
wage wo k in ag icul u e o household heads should be obse ed. This
could impac households’heads engagemen in own-accoun ag icul u e
and po en ially ag icul u al yields, bu he di ec ion o his impac is
unce ain. Addi ional income could in luence acquisi ion o inpu s and
po en ial highe yields; o new wage-ag icul u al employmen could
d i e-ou own-accoun ag icul u al ac i i y. A ailable e idence sugges s
non a m income inc eases a m hi ed labo and dec eases ag icul u al
p oduc i i y, as a me s use i o mo e ou o ag icul u e a he han
in es ing in c op p oduc ion (Kilic e al., 2009; Ama e and Shi e aw,
2017).
The hi d pa hway ep esen ed in Figu e 1 shows ha o es y
plan a ions may also induce a m displacemen . I a me s we e pushed
o ma ginal lands, an inc ease in cul i a ed a ea may be obse ed, bu no
changes in c op a ie ies ( adi ional c ops should s ill be p edominan ).
Because ma ginal lands would p esumably be o lowe soil quali y,
yields should dec ease, and a m employmen dec ease as he a me has
lowe capi al o hi e wo ke s. No changes in he household heads’
employmen sec o would be expec ed. On he o he hand, i a me s
we e pushed o o he economic sec o s, a decline in cul i a ed a ea
should be expec ed and a change in he employmen sec o o household
head, wi h a dec ease in he likelihood o ha ing ag icul u e as he main
occupa ion and an inc ease in non a m employmen (ei he wage o sel -
employmen ) o e en in unemploymen o inac i i y.
Gi en ha o es y plan a ions in Mozambique a e mainly mono-
cul u es ocused on pine and eucalyp us, which a e no labo in ensi e
cul i a ions, we hypo hesize ha he second pa hway, di ec employ-
men c ea ion, will no be s ong. Since in es men s in o o es y is a
new phenomenon and land dema ca ed o o es y o en o e laps wi h
communi y land inhabi ed o used o li elihood ac i i ies, we hypo h-
esize ha he hi d pa hway in Figu e 1 is likely o domina e mo e han
he i s pa hway. Since a combina ion o a push o ma ginal lands o
o he economic sec o s is bo h likely, i is unce ain whe he he cul i-
a ed a ea would inc ease o dec ease; bu i is expec ed ha a me s
may emain cul i a ing adi ional c ops, achie ing low yields, and
main aining none o minimum le els o hi ed a m labo . I is also un-
ce ain whe he a me s would emain as own-accoun a m wo ke s o
be pushed ou o ag icul u e, ei he in o he non a m sec o o in o
unemploymen o inac i i y.
4. Me hodology
4.1. Da a
This s udy elies on wo p ima y sou ces o da a. Fi s , he ex en o
o es y plan a ions and hei expansion ajec o ies a e ob ained om
emo e sensing algo i hms ha p ocess Uni ed S a es Geological Su ey
(USGS) Landsa 7 and Landsa 8 image y o he yea s 2001, 2006, 2012,
and 2017, as desc ibed in Bey and Mey oid (2021). The a ailable
da ase maps o es y plan a ions o each yea , and he changes in he
ype o land use o e ime be ween 2001 and 2017. Such changes
include wo ypes o land use ajec o ies, depending on he p e ious
land use/co e be ween 2001 and 2017 whe e he ee plan a ions go
es ablished, i.e., p e ious land use/co e being na u al ege a ion
(which includes mosaics o o es s, woodlands and g asslands), e sus
being c opland (Bey and Mey oid , 2021). To dis inguish la ge-scale
o es y plan a ions om small-scale ope a ions, his wo k excluded
all woodlo s o i e ha o less. The o es y in es men s iden i ied
co espond essen ially o ee plan a ions such as pine and eucalyp us,
bu also o macadamia, mango, and ci us, o a lesse ex en . The da a
only includes plan a ions which we e newly es ablished o expanded
a e he yea 2001.
We combine his in o ma ion wi h he second da a sou ce, i.e., wo
c oss-sec ional na ionally ep esen a i e In eg a ed Ag icul u al Su eys
(he ea e e e ed o as IAI om i s Po uguese ac onym), o he yea s
3
Degne e al. (2022) s udy he ce i ica ion o o es y plan a ions in he
con ex o Mozambique, which in ol es a ma ke -d i en, non-s a e go e nance
sys em aimed a p omo ing sus ainable o es managemen (SFM). This ce i-
ica ion seeks o incen i ize o es owne s o adhe e o SFM s anda ds by o -
e ing inancial o epu a ional ewa ds, such as p ice p emiums and enhanced
ma ke access o ce i ied p oduc s. I is p ima ily implemen ed h ough
schemes like he Fo es S ewa dship Council (FSC), which se s c i e ia o
esponsible o es managemen , including p inciples add essing communi y
igh s and ela ions wi hin he managemen a ea. The ce i ica ion p ocess is
in ended o imp o e social aspec s, including in e ac ions be ween plan a ion
owne s and local communi ies.
C. Chia ella e al.
Land Use Policy 145 (2024) 107251
4
2007 and 2017. The su eys a e adminis e ed by he Minis y o Ag i-
cul u e and Ru al De elopmen in close pa ne ship wi h he Na ional
Ins i u e o S a is ics. The yea 2007 is chosen as he baseline pe iod as i
coincides wi h he s a o he newes in es men wa e (which includes
o es y in es men s), he a ailabili y o IAI da a, and he emo ely
sensed maps o o es y plan a ions (Bey and Mey oid , 2021). The yea
2017 is well pas he end o he majo o es y in es men wa e and
complemen a y da a o he IAI and o he emo ely sensed maps is
a ailable o his yea . Thus, 2017 se es as he pos - ea men yea .
These su eys con ain GPS coo dina es o household loca ions and he
names o he adminis a i e a eas down o he communi y le el. The
su ey includes de ailed modules on demog aphic cha ac e is ics,
plan ing and ha es ing decisions a he plo and c op le el, labo use
and ea nings, asse owne ship, sales, and ood secu i y.
We o e lap he household loca ions
4
om 2007 and 2017 wi h he
loca ion o he plan a ions o each o he e alua ed ounds, as well as
hei expansion ajec o ies as illus a ed in Figu e 2. This igu e shows
he ou p o inces whe e he s udy akes place in no he n Mozambique,
he GPS loca ions o households su eyed in yea s 2007 and 2017, and
zooms in he o es y plan a ions ajec o ies be ween 2001 and 2017. In
o ange is he ajec o y co esponding o plan a ions ha expanded on
c opland, and in pu ple he ajec o y co esponding o plan a ions ha
expanded on na u al ege a ion. The ime be ween 2007 and 2017
co esponds o he inc ease in o eign o es y in es men s in he a ea,
which began a ound 2008 and peaked in 2012 (Bey and Mey oid ,
2021). Acco dingly, he mapped plan a ion a ea inc eased om 4983 ha
in 2006, o 7832 ha in 2012, o 18,178 ha in 2017, wi h 70 % o he
plan a ions expanding on p e ious ag icul u al land (Bey and Mey oid ,
2021).
4.2. Empi ical s a egy
The s udy co e s he a ming households in he p o inces o Nam-
pula, Niassa, Zambezia and Cabo Delgado, which is he co e age o he
emo e sensing-based maps o o es y plan a ions. We limi he analysis
o hose households wi hin 250 km o any o es y plan a ion, o exclude
con ex s ha a e expec ed o be oo di e en om hose a ec ed by
plan a ions (consis en wi h Deininge and Xia, 2016).
We e alua e he impac s o exposu e o o es y plan a ions on
neighbo ing smallholde s’p oduc i e and employmen ou comes, using
a Di e ence-in-Di e ence (DID) app oach. To implemen he DID, we
cons uc a sample o households exposed (i.e., ea ed) and unexposed
(i.e., con ol) o o es y plan a ions. We de ine exposed households as
hose ha ing a minimum sha e o 0.1 % a ea occupied wi h o es y
plan a ions wi hin a ci cula bu e o adius a ound he household in
he pos - ea men pe iod (i.e., 2017). Since he IAI did no e isi he
same a ming households in bo h yea s (2007 and 2017), we use he
da a om households loca ed wi hin he same a eas subsequen ly
a ec ed, o eplace he p e-exposu e p oduc i e and employmen mea-
su es, i.e., p e-exposu e ea ed households a e hose ha ing a minimum
0.1 % a ea occupied wi h pos - ea men o es y plan a ion wi hin a
ci cula bu e o adius.
5
This ensu es ha households in he 2007 p e-
exposu e g oup a e loca ed in simila a eas as hose in 2017 pos -
exposu e g oup.
We chose o de ine bo h exposed and un-exposed households in
ela ion o o es y plan a ions in he pos - ea men pe iod delibe a ely
o adhe e o he ’no composi ional di e ences’assump ion o he DID
model. By selec ing ea ed households based on hei p oximi y o he
2017 plan a ions, ou in en is o ensu e consis en de ini ion o ea ed
households in bo h p e and pos pe iods, si ua ing hem wi hin simila
plan a ion-sui able a eas. This app oach is simila o a eg ession
Fig. 1. Possible pa hways o plan a ions impac s.
4
Se en y se en pe cen o he sampled households in 2007 we e missing GPS
coo dina es. We hence inpu ed he GPS coo dina es o he P ima y Sampling
Uni s (PSU) cen oids o hese households. Eigh a ming households we e
sampled pe PSU.
5
This is, exposed households in he p e-exposu e pe iod (i.e., 2007) a e hose
ha ing a ea occupied wi h 2017 o es y plan a ions wi hin a ci cula bu e o
adius a ound he household.
C. Chia ella e al.
Land Use Policy 145 (2024) 107251
5
discon inui y design, in ha i in ol es designa ing households jus
ou side he ’a ea o exposu e’as con ols, dema ca ing and capping he
ex en o his exposu e a ea. Hence, ea ed and con ol g oups a e
compa able no jus in p oximi y bu also in obse able and unobse -
able cha ac e is ics. While his app oach may no en i ely elimina e
composi ional di e ences inhe en in epea ed c oss-sec ions, i p o-
ides a mo e obus compa ison in he absence o panel da a, ensu ing
no composi ional di e ences be ween g oups.
Many easons explain he es ablishmen o o es y in es men in
ce ain a eas, such as a ailable land, app op ia e slope, e ile soils,
among many o he s. Households loca ed in he icini y o such o es y
in es men s may hence di e om households loca ed u he away, a
di e ence known as selec ion bias. The e is addi ionally a ime a ian
componen . Ou come a iables o all households change o e ime
ega dless o hei exposu e o o es y in es men s, a componen known
as a ime end. When he selec ion bias is imely in a ian and he ime
end is he same o bo h exposed and unexposed g oups, a pa allel
ends o common ends assump ion holds and we can causally e alua e
he e ec s o o es y in es men using he DID app oach. Gi en ha we
compa e households ha a e wi hin 250 km o o es y plan a ions, in
he same a eas o he ou p o inces, we conside he la e wo condi-
ions easonable, as exposed households a e no oo dis an om con ol
households. As a obus ness check, we also a y he de ini ion o such
exposu e h eshold h ough a sensi i i y analysis. Equa ion (1) de ails
he DID s a egy.
Yip =β+γExpip +λPos ip +δExpip ×Pos ip +Xipκ+
α
p+ϵip (1)
Whe e iis a subsc ip o each household and pa subsc ip o each
p o ince.
α
p
a e p o ince ixed e ec s, which we include o accoun o
possible egional he e ogenei ies. Pos is he ime end which equals one
o households su eyed a he endline pe iod in 2017 and ze o o hose
su eyed a he baseline pe iod in 2007. Exp indica es he exposu e o
o es y plan a ions. Fo ou basic speci ica ion we de ine Exp as a
‘p esence’indica o a iable ha akes he alue o one i a leas 0.1 %
o he a ea o a concen ic ci cle a ound each household is occupied wi h
o es y plan a ions es ablished a e 2007. We choose a h eshold o
0.1 % o he a ea a ound each household as he indica o o he ‘p es-
ence’o a plan a ion, as opposed o a bina y a iable indica ing he
p esence o a plan a ion in he concen ic ci cle ega dless o plan a ion
a ea, which may lead o he inclusion o households as ea ed, despi e
negligible a ea occupied wi h plan a ions a ound he household. This
adds an ex a laye o obus ness by p e en ing small classi ica ion o
loca ion e o s. This basic exposu e a iable allows us o es ima e a
s anda d DID model wi h a bina y ‘ ea men ’ a iable, be o e and a e
he expansion o he plan a ions. This speci ica ion conside s as ‘con ol’
o coun e ac ual g oup all hose households wi h less han 0.1 % o he
a ea o he concen ic bu e a ound he household occupied wi h
o es y plan a ions, no ing ha he majo i y o households ha e no
plan a ion p esence a all.
X
ip
is a ec o o household cha ac e is ics, in which we include de-
mog aphic cha ac e is ics o he household head (sex, age and yea s o
educa ion), and whe he he head o household wo ks as an employee o
is sel -employed.
6
ϵ
ip
a e andom dis u bances, which a e independen
and iden ically dis ibu ed N(0,
σ
2
). S anda d e o s a e clus e ed a he
dis ic le el. Y
ip
co esponds o he ou come a iables, which a e: i) a
se o p oduc i e ou comes: a m size in ha ( o al a ea sel - epo ed by
he household, cul i a ed land size, a ea unde pe manen c ops, and
o al a ea measu ed by enume a o s), he alue o all c ops p oduced (in
2017 USD PPP), o al ag icul u al yields ( he alue o all c ops p oduced
Fig. 2. Mapping o o es y plan a ions ajec o ies and households su eyed in 2007 and 2017.
6
We conside as con ols he head o household employmen s a us o
p oduc i e ou comes and c op choices bu no o employmen ou comes.
C. Chia ella e al.
Land Use Policy 145 (2024) 107251
6
in 2017 USD PPP pe ha), he alue o all c ops sold (in 2017 USD PPP),
he alue o such sales pe ha, and he o al cos o seeds (in 2017 USD
PPP); ii) a se o employmen ou comes: a m employmen (whe he he
household employed wo ke s ull ime, pa - ime, he o al numbe o
wo ke s employed, and numbe o men and women employed),
dicho omous a iables o whe he he head o household wo ked as an
employee o sel -employed, dummy a iables o whe he ag icul u e
was he main ac i i y, a seconda y ac i i y o he head o household did
no p ac ice ag icul u e; and iii) a se o c op choice ou comes: dummy
a iables o whe he he household cul i a ed he main c ops cul i a ed
in he a ea, maize, ice, so ghum, g oundnu s, beans, and sesame.
The pa ame e o in e es in he abo e speci ica ion is δ, he coe i-
cien o he in e ac ion be ween exposu e and pos s a us, which cap-
u es he impac o exposu e o plan a ions on he ou comes o in e es a
he end o he pe iod (2017). I es ima es he A e age T ea men E ec
on he T ea ed (ATT), o hose households ha we e exposed o he
appea ance o o es y plan a ions a e 2007. By accoun ing o a
double di e ence, he DID app oach can dis inguish be ween e ec s o
he ea men i sel (caused by he exposu e o plan a ions) and o he
ime-dependen ac o s ha may be a ec ing he ou come a iable in
bo h exposed and unexposed a eas.
The delimi a ion o he adius o in luence o he exposu e a iable
is a sensi i e decision. To ob ain esul s ha a e obus o unce ain ies
in he ac ual dis ance up o which plan a ions migh a ec households,
we i s conside a adius o in luence o 20 km a ound each household,
and conduc sensi i i y analysis a ying he adius o exposu e o: 5, 10,
15, 25 and 30 km a ound each household.
Wha cons i u es exposu es o plan a ions is ano he sensi i e deci-
sion. Plan a ion exposu e can ake he o m o he me e p esence o a
plan a ion in a households’ icini y, bu i may also be conside ed as
s onge o weake depending on he p oximi y o he household o he
plan a ion (dis ance), and on how much a ea he plan a ion o g oup o
plan a ions occupies in he su ounding landscape (in ensi y o expo-
su e). To es o hese measu es o in ensi y, we conside wo addi ional
indica o s: i) a con inuous a iable wi h he in e se o he dis ance o he
closes poin whe e a o es y plan a ion was es ablished a e 2007; and
ii) a con inuous a iable wi h he sha e o he a ea o a concen ic ci cle
a ound each household occupied wi h o es y plan a ions es ablished
a e 2007. We e alua e bo h in ensi y a iables h ough he ollowing
DID speci ica ion:
Yip =β+γExpip +λPos ip +δExpip ×Pos ip ×In ip +Xipκ+
α
p+ϵip (2)
whe e In co esponds o he in ensi y a iables desc ibed abo e. Tes ing
hese con inuous a iables may esul impo an in he case o e alu-
a ing o es y in es men s, as we may ca e mo e abou he e ec o he
changes in he in ensi y o exposu e (in ensi e ma gin) han abou he
exis ence o he o es y in es men s (ex ensi e ma gin).
To u he unde s and he mechanisms explaining he impac s, we
disen angle he exposu e a iable by he expansion ajec o ies. We
conside h ee possible cases depending on p e ious land uses, i.e., all
plan a ions in he bu e zone we e es ablished in na u al ege a ion, all
plan a ions in he bu e zone we e es ablished in c opland, and plan-
a ions in he bu e zone we e es ablished bo h in na u al ege a ion
and c opland a eas. As he basic speci ica ion, we conside a household
-‘exposed’i a o es y plan a ion om one o hese ajec o ies occupied
an a ea g ea e han 0.1 % o he 20 km bu e zone a ound each
household. Wi h his in o ma ion, we es ima e equa ion (3).
whe e NV and CL co espond o na u al ege a ion and c opland a-
jec o ies, and NVCL o a household ha expe ienced he es ablishmen
o bo h ajec o ies in he icini y. δ
1
,δ
2
, and δ
3
es ima e he ATT e ec s
o each o he ajec o ies and hei in e ac ion.
5. Resul s
In his sec ion we i s p esen desc ip i e s a is ics o he exposu e,
ou come, and con ol a iables. We hen show he esul s o he es i-
ma ions o exposu e o o es y plan a ions, by g ouping ou comes in
h ee g oups: p oduc i e ou comes ( a m size, yields, sales), employ-
men ou comes (hi ed employmen and he sec o o employmen o he
household head), and c op choices (whe he he household a ms he
main c ops g own in he a ea). We i s p esen he esul s o he basic
speci ica ion, which e alua es he ex ensi e ma gin o exposu e o
plan a ions. We also conduc he sensi i i y analysis on he chosen 20-
km bu e , o e alua e how esponsi e a e he e ec s o he choice o
he bu e adius. We hen e alua e he in ensi e ma gin, by es ima ing
he e ec s o dis ance o o es y in es men s and he ex ension o he
occupied a ea. Finally, we show he e ogeneous impac s o he expan-
sion ajec o ies.
5.1. Desc ip i e s a is ics
Fi s , we p esen desc ip i e s a is ics o he exposu e indica o s o
plan a ion p esence, sha e o plan a ion in each bu e ing, and o dis-
ance o nea es plan a ion, o p e-and pos -exposu e samples (Table 1).
As a eminde , we de ine hese ea men o exposu e a iables ela i e
o he plan a ions a endline (2017) also o p e-exposu e households.
Hence, Table 1 shows how balanced in loca ion a e p e and pos -
exposu e samples. Abou 4 % o households in bo h p e- and pos -
exposu e pe iods obse ed a plan a ion wi hin a 20 km bu e adius
a ound he household. The sha e a ies sligh ly o di e en adii o he
concen ic ci cles, bu he e a e no signi ican di e ences ac oss he
pe iods. Households’exposu e o plan a ions, i,e., in e ms o bo h he
p esence o plan a ions and he sha e o plan a ions, a a 5 km bu e
adius is highe in he p e-exposu e pe iod. Beyond he 5 km adius,
exposu e o plan a ions shows no signi ican di e ence ac oss p e-and
pos -exposu e pe iods. Bu households in pos -exposu e sample a e
signi ican ly close o plan a ions han hose in he p e-exposu e sample.
Table 2 shows he mean di e ences in ou come a iables and
co a ia es ac oss exposed and con ol households, o be o e and a e
exposu e. The me ics show ha p io o exposu e, he sample o
exposed households we e mo e likely o ely on ag icul u e as hei
p ima y li elihood ac i i y han he unexposed con ol households
7
: i.e.,
in he p e-exposu e pe iod, he exposed households we e likely o wo k
la ge a ms bo h in e ms o o al a ea and cul i a ed a ea, hi e mo e
ull- ime wo ke s, cul i a e maize and so ghum, and incu ed highe
seed cos s. These exposed households we e also likely o ha e a
dependen sou ce o income (sala y) and be headed by emales han
Yip =β+ +λPos ip +γ1ExpNVip +γ2ExpCLip +γ3ExpNVCLip+
δ1ExpNVip ×Pos ip +δ2ExpCLip ×Pos ip +δ3ExpNVCLip ×Pos ip +Xipκ+
α
p+ϵip (3)
7
i.e., compa ing he unexposed and exposed columns in he le panel (o
P e-pe iod) o Table 2
C. Chia ella e al.
Land Use Policy 145 (2024) 107251
7
unexposed households. The unexposed households on he o he hand
we e mo e likely o ha e ag icul u e as a seconda y ac i i y, cul i a e
ice and g oundnu s, and o be headed by a pe son wi h mo e yea s o
educa ion. Exposed and unexposed samples p io o exposu e a e
balanced in o al a m a ea ha was measu ed by enume a o s, o al
alue o he ou pu , he sale alue o he ou pu , he o al numbe o
wo ke s, he numbe o pa - ime wo ke s, whe he he wo ke s we e
male o emale, he cul i a ion o beans, and he age o he household
head.
These di e ences a p e-exposu e should no cons i u e a p oblem o
iden i ica ion as long as he exposed and con ol a eas ha e ollowed
pa allel ends. Ideally, we would ha e epea ed p e-exposu e measu es
o check o pa allel ends,
8
bu as Ro h (2022) sugges , gi en limi a-
ions in he p ac ice o es ing o p e- ends, using con ex -speci ic
knowledge o discuss possible iola ions o pa allel ends will yield in
mo e c edible in e ence. We know ha 70 % o plan a ion expansion
occu s on c opland a he han on na u al ege a ion. I has been shown
also ha communi y lands and p oximi y o p io s a e a ms a e he
main d i e s o plan a ion expansion (Bey, 2021). Since such a e he
cha ac e is ics o he a eas whe e mos sampled a ms a e es ablished,
Table 1
Mean di e ences p e e sus pos on plan a ion exposu e a iables.
P e Pos
Mean S.D. Obs. Mean S.D. Obs. Di .
P esence o o es y plan a ion a 5 km (Yes/No) 0.032 0.18 2106 0.019 0.14 2753 −0.01***
P esence o o es y plan a ion a 10 km (Yes/No) 0.036 0.19 2106 0.034 0.18 2753 −0.00
P esence o o es y plan a ion a 15 km (Yes/No) 0.036 0.19 2106 0.034 0.18 2753 −0.00
P esence o o es y plan a ion a 20 km (Yes/No) 0.036 0.19 2106 0.037 0.19 2753 0.00
P esence o o es y plan a ion a 25 km (Yes/No) 0.036 0.19 2106 0.044 0.21 2753 0.01
P esence o o es y plan a ion a 30 km (Yes/No) 0.041 0.20 2106 0.050 0.22 2753 0.01
Sha e o plan a ion a 5 km bu e (%) 0.002 0.01 2106 0.001 0.01 2753 −0.00***
Sha e o plan a ion a 10 km bu e (%) 0.001 0.01 2106 0.001 0.01 2753 −0.00
Sha e o plan a ion a 15 km bu e (%) 0.001 0.01 2106 0.001 0.01 2753 0.00
Sha e o plan a ion a 20 km bu e (%) 0.001 0.00 2106 0.001 0.00 2753 0.00
Sha e o plan a ion a 25 km bu e (%) 0.001 0.00 2106 0.001 0.00 2753 0.00
Sha e o plan a ion a 30 km bu e (%) 0.001 0.00 2106 0.001 0.00 2753 0.00
Dis ance o nea es 2017 plan a ion 2391.410 2136.36 2106 1911.937 1583.34 2753 −479.47***
*p<0.10. **p<0.05, *** p<0.01.
Di . column shows di e ence be ween 2007 and 2017 means
Table 2
Mean di e ences by exposu e s a us on ou come and con ol a iables p e and pos exposu e.
P e Pos
Unexposed Exposed Di . Unexposed Exposed Di .
P oduc i e ou comes
To al a ea sel - epo ed (ha) 1.66 2.02 0.36** 1.32 2.30 0.98***
Cul i a ed a ea (ha) 1.56 2.00 0.45*** 1.22 2.13 0.91***
A ea wi h pe manen c ops (ha) 0.02 0.00 −0.02 0.01 0.00 −0.01
To al a ea as measu ed by enume a o s (ha) 1.51 1.63 0.12 1.23 1.23 0.00
Value o ou pu o all c ops, 2017 USD PPP 87.07 135.89 48.82 334.40 380.10 45.70
Yields o all c ops, 2017 USD PPP pe HA 53.27 56.76 3.49 356.21 278.49 −77.71
Sales alue o all c ops, 2017 USD PPP 26.38 48.12 21.74 198.15 166.15 −32.00
Sales pe ha o all c ops, 2017 USD PPP pe HA 14.52 18.87 4.35 373.19 160.34 −212.85
Seeds cos o all c ops, 2017 USD PPP 0.60 3.97 3.37*** 4.45 13.41 8.96***
Employmen ou comes
Hi ed wo ke s ull- ime (Yes/No) 0.02 0.05 0.03* 0.02 0.08 0.05***
Hi ed wo ke s pa - ime (Yes/No) 0.20 0.26 0.07 0.14 0.16 0.02
Numbe o o al wo ke s hi ed (N) 6.51 5.50 −1.01 3.79 5.43 1.64
Hi ed male wo ke s (N) 4.13 4.00 −0.13 2.98 3.43 0.45
Hi ed emale wo ke s (N) 2.38 1.50 −0.88 0.80 2.00 1.20*
H head wo ks as wage wo ke (Yes/No) 0.24 0.36 0.12** 0.23 0.23 −0.01
HH head wo ks as sel -employed (Yes/No) 0.52 0.48 −0.04 0.45 0.48 0.03
Ag icul u e is main ac i i y o HH head (Yes/No) 0.82 0.93 0.11** 0.88 0.90 0.03
Ag icul u e is seconda y ac i i y o HH head (Yes/No) 0.13 0.07 −0.07* 0.09 0.05 −0.03
HH head does no p ac ice ag icul u e (Yes/No) 0.04 0.00 −0.04* 0.04 0.04 0.01
C op choices
Cul i a ed maize (Yes/No) 0.68 0.99 0.31*** 0.74 0.89 0.15***
Cul i a ed ice (Yes/No) 0.27 0.04 −0.24*** 0.15 0.06 −0.09**
Cul i a ed so ghum (Yes/No) 0.29 0.54 0.25*** 0.18 0.30 0.13***
Cul i a ed g oundnu s (Yes/No) 0.41 0.28 −0.13** 0.44 0.18 −0.27***
Cul i a ed beans (Yes/No) 0.62 0.71 0.09 0.62 0.61 −0.01
Cul i a ed sesame (Yes/No) 0.00 0.00 0.00 0.06 0.01 −0.05**
Con ol a iables
HH head emale 0.20 0.32 0.11** 0.27 0.28 0.01
HH head age 41.15 41.91 0.76 42.09 42.53 0.45
HH head yea s o educa ion 2.77 1.79 −0.99*** 3.68 3.91 0.24
*p<0.10,**p<0.05, ***p<0.01.
Di . column shows he mean di e ence and signi icance be ween households exposed and unexposed o o es y plan a ions, o 2007 and 2017
8
al hough gi en ha we do no ollow he same households o e ime, his
would s ill be an impe ec p oxy
C. Chia ella e al.
Land Use Policy 145 (2024) 107251
8
and since ag icul u al income and employmen ou comes ha e been
ai ly s able ac oss exposed and con ol a eas as hese a e p edominan ly
subsis ence households ha ha e been wo king in a ming o yea s, i is
easonable o assume ha he ou comes o in e es ha e e ol ed in
pa allel p io o exposu e ac oss hese a eas.
Di e ences be ween he exposed and unexposed households
con inue o e o he pos -exposu e pe iod, excep o he ollowing: i.e.,
in he pos -exposu e pe iod, he exposed households end o hi e emale
wo ke s and he unexposed households end cul i a e sesame, while
p e ious di e ences ha exis ed in o he employmen ( a m o non-
a m), main income ac i i y, household head’s pa icipa ion in
ag icul u e, and household head being emale and ha ing mo e yea s o
educa ion end o disappea .
5.2. Impac s o exposu e in e ms o he p esence o plan a ions
Table 3 shows he es ima ed impac s o exposu e o plan a ions on
households’p oduc i e ou comes. Exposed households expe ience an
inc ease in he sel - epo ed o al a m a ea o 0.6 ha, which is la ge
gi en he a e age size o a ms in he sample. Howe e , no signi ican
di e ence is de ec ed in cul i a ed a ea, a ea wi h pe manen c ops, and
o al a m a ea as measu ed by enume a o s. Exposu e o plan a ion also
Table 3
Di -in-Di e ec s o exposu e based on plan a ion sha e on p oduc i e ou comes.
A ea o al
(Ha)
A ea cul i a ed
(Ha)
A ea Pe m
(Ha)
A ea measu ed
(Ha)
Ou pu alue
(PPP)
Yields (PPP/
Ha)
Sales alue
(PPP)
Sales/Ha
(PPP/Ha)
Seeds cos
(PPP)
Dummy exposu e 0.05 0.14 −0.00 −0.17 −294.27 −142.08 −50.40* −24.48 3.77***
(0.17) (0.18) (0.01) (0.19) (191.71) (119.92) (27.47) (65.82) (1.41)
Dummy pos −0.35*** −0.34*** −0.00 −0.26** 236.24*** 289.97*** 174.92*** 385.65 3.62***
(0.09) (0.08) (0.01) (0.12) (48.17) (42.78) (57.60) (270.17) (0.45)
Exposu eXPos 0.58* 0.44 0.00 −0.15 36.17 −80.79 −33.28 −74.70 5.07
(0.32) (0.32) (0.01) (0.17) (75.94) (61.32) (46.40) (121.93) (8.11)
Female HHhead −0.39*** −0.37*** −0.01** −0.52*** −123.18*** −50.69** 68.27 603.61 0.02
(0.05) (0.04) (0.01) (0.11) (28.14) (24.07) (129.54) (640.38) (0.84)
Age HHhead 0.01*** 0.01*** 0.00** 0.01*** 0.91 −0.02 −1.64** −3.90 0.01
(0.00) (0.00) (0.00) (0.00) (0.59) (0.70) (0.75) (3.10) (0.02)
Yea s o Edu HHhead 0.01 0.01 0.00 −0.03 3.70 6.52 −3.38 −28.06 0.35***
(0.01) (0.01) (0.00) (0.02) (4.19) (3.94) (5.91) (26.79) (0.13)
Dummy wage employed −0.01 −0.01 −0.01 0.11 12.85 30.45 105.41 628.60 0.58
(0.07) (0.07) (0.01) (0.16) (27.96) (26.72) (123.51) (623.06) (0.68)
Dummy sel employed 0.11** 0.08* 0.01 −0.07 25.12 14.59 106.77 378.26 1.58***
(0.05) (0.05) (0.01) (0.11) (21.68) (25.15) (69.70) (340.77) (0.57)
P o ince ixed e ec s ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
R
2
0.05 0.05 0.00 0.04 0.07 0.06 0.00 0.00 0.03
N 4637 4637 4637 1118 4763 4621 4664 4572 4664
Mean o dep. a con ol
a baseline
1.66 1.56 0.02 1.51 87.07 53.27 26.38 14.52 0.60
No es: S anda d e o s in pa en heses
*p<0.10, **p<0.05, ***p<0.01.
Table 4
Di -in-Di e ec s o exposu e based on plan a ion sha e on employmen ou comes.
Employed
ull ime
(Yes/No)
Employed
pa ime
(Yes/No)
Wo ke s
employed
(N)
Male
wo ke s
employed
(N)
Female
wo ke s
employed
(N)
Wage
Emp
(Yes/No)
Sel Emp
(Yes/No)
Ag
MainAc
(Yes/No)
Ag Seconda y
Ac (Yes/No)
Non a m
(Yes/No)
Dummy exposu e −0.02 0.03 1.68* 0.86 0.82 0.18** 0.03 0.04 −0.03 −0.01
(0.02) (0.07) (0.99) (0.64) (0.93) (0.08) (0.06) (0.03) (0.02) (0.01)
Dummy pos −0.01 −0.07*** −2.25* −0.88 −1.36* −0.02 −0.05*** 0.08*** −0.07*** −0.02
(0.01) (0.02) (1.16) (0.68) (0.80) (0.02) (0.02) (0.02) (0.01) (0.01)
Exposu eXPos 0.03 −0.07 0.52 −0.56 1.08 −0.16** 0.03 −0.04 −0.00 0.04*
(0.04) (0.09) (1.67) (0.65) (1.31) (0.08) (0.06) (0.03) (0.02) (0.02)
Female HHhead −0.01** −0.05*** −1.51 −0.62 −0.89 −0.07*** −0.18*** 0.01 −0.03*** 0.02**
(0.00) (0.01) (1.07) (0.69) (0.71) (0.01) (0.02) (0.01) (0.01) (0.01)
Age HHhead 0.00*** 0.00*** −0.04 −0.02 −0.02 −0.00*** −0.00*** −0.00** 0.00 0.00**
(0.00) (0.00) (0.03) (0.02) (0.02) (0.00) (0.00) (0.00) (0.00) (0.00)
Yea s o Edu
HHhead
0.00*** 0.02*** 0.04 0.09 −0.05 0.02*** 0.01*** −0.04*** 0.03*** 0.01***
(0.00) (0.00) (0.10) (0.07) (0.05) (0.00) (0.00) (0.00) (0.00) (0.00)
P o ince ixed
e ec s
✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
R
2
0.02 0.05 0.14 0.12 0.13 0.05 0.05 0.12 0.09 0.03
N 4763 4763 111 111 111 4763 4763 4763 4763 4763
Mean o dep.
a con ol a
baseline
0.02 0.20 6.51 4.13 2.38 0.24 0.52 0.82 0.13 0.04
No es:S anda d e o s in pa en heses
*p<0.10, **p<0.05, ***p<0.01.
C. Chia ella e al.
Land Use Policy 145 (2024) 107251
15
MozLegal (2004). Land law legisla ion. h p://www.impac o.co.mz/impac o-en/wp-
con en / hemes/A po a2_1_0/pd /Lei%20de%20Te as/Ing/Mozambique-Land-
Law-Legisla ion.pd .
Mülle , M.F., Penny, G., Niles, M.T., Riccia di, V., Chia elli, D.D., Da is, K.F., e al., 2021.
Impac o ansna ional land acquisi ions on local ood secu i y and die a y di e si y.
P oc. Na l. Acad. Sci. 118 (4), e2020535118.
Muyanga, M., Jayne, T.S., 2019. Re isi ing he a m size-p oduc i i y ela ionship based
on a ela i ely wide ange o a m sizes: e idence om Kenya. Am. J. Ag ic. Econ.
101 (4), 1140–1163.
Nhan umbo, I., Macqueen, D., C uz, R., and Se a, A.F.A. (2013). In es ing in locally
con olled o es y in Mozambique: Po en ial o p omo ing sus ainable u al
de elopmen in he p o ince o Niassa. In e na ional Ins i u e o En i onmen and
De elopmen . h ps://www.iied.o g/si es/de aul / iles/pd s/mig a e/13569IIED.pd
.
Nol e, K., Os e meie , M., 2017. Labou ma ke e ec s o la ge-scale ag icul u al
in es men : concep ual conside a ions and es ima ed employmen e ec s. Wo ld
De . 98, 430–446.
No a, Y. and Ros´
a io, R. (2022). Ag icul u al de elopmen models in mozambique.
Obse a o io do Meio Ru al, (129). h ps://om mz.o g/wp-con en /uploa
ds/2022/09/OR-129-Ag icul u al-de elopmen -models-in-Mozambique-1.pd .
Obe lack, C., Gige , M., Anseeuw, W., Adelle, C., Bou blanc, M., Bu nod, P., Ecke , S.,
Fi awek, W., Fouilleux, E., Hend iks, S., e al., 2021. Why do la ge-scale ag icul u al
in es men s induce di e en socio-economic, ood secu i y, and en i onmen al
impac s? E idence om kenya, Madagasca , and Mozambique. Ecol. Soc. 26 (4)
a icle–18.
Oli ei a, E., Mey oid , P., 2022. S a egic spa ial planning in eme ging land-use
on ie s: E idence om Mozambique. Ecol. Soc. 27, 2.
O lowski, M. (2016). On he w ong ack?! o es plan a ions in mozambique.
Koo dinie ungsk eis Mosambik, Biele eld, Ge many. h ps://s a ic1.squa espace.
com/s a ic/5ede6 5665eb9856c72471b2/5 76584a9881444e9783b 35/5 7659
489881444e9783e9cd/1601591624763/KKM_OnTheW ongT ack.pd ? o ma
=o iginal.
Phimma ong, S., Keenan, R.J., 2020. Fo es plan a ion de elopmen , po e y, and
inequali y in laos: a dynamic cge mic osimula ion analysis. Fo . Policy Econ. 111,
102055.
Rasmussen, M.B., Lund, C., 2018. Recon igu ing on ie spaces: he e i o ializa ion o
esou ce con ol. Wo ld De . 101, 388–399.
Ro h, J., 2022. P e es wi h cau ion: e en -s udy es ima es a e es ing o pa allel
ends. Am. Econ. Re .: Insigh s 4 (3), 305–322.
Se zedelo de Almeida, L., Delgado, C., 2019. The plan a ion o es y sec o in
mozambique: Communi y in ol emen and jobs. Wo ld Bank, Washing on, DC.
h ps://doi.o g/10.1596/31753.
Si ko, N.J., Chisanga, B., Tschi ley, D., Jayne, T.S., 2018. An e olu ion in he middle:
examining he ise o mul ina ional in es men in smallholde g ain ading in
zambia. Food Secu . 10, 473–488.
Si oe, A. and Lisboa, S.N. (2020). A aliaç˜
ao dos impac os dos in es imen os nas
plan aç˜
oes lo es ais da po ucel-moçambique na p o íncia da zamb´
ezia.
Obse a o io do Meio Ru al, (88).
Si oe, A., Salom˜
ao, A., We z-Kanounniko , S., e al. (2012). The con ex o edd+in
mozambique: causes, ac o s and ins i u ions.CIFOR Occasional Pape , (76).
h ps://webdoc.sub.gwdg.de/ebook/se ien/yo/CIFOR_OP/79.pd .
Sulli an, J., B own, D., Moyo, F., Jain, M., Ag awal, A., 2022. Impac s o la ge-scale land
acquisi ions on smallholde ag icul u e and li elihoods in anzania. En i on. Res.
Le . 17 (8), 084019.
Ve meulen, S., Co ula, L., 2010. O e he heads o local people: consul a ion, consen ,
and ecompense in la ge-scale land deals o bio uels p ojec s in a ica. J. Peasan
S ud. 37 (4), 899–916.
Wo ld Bank (2016). Republic o mozambique: Imp o ing he business clima e o plan ed
o es s, inal epo . Wo ld Bank, Washing on, DC, (Repo No. ACS18952. A ailable
a :: 〈h ps://www.p o o .in o/si es/p o o .in o/ iles/FinalRepo Imp o ingIn
es men Clima eMoz_0.pd )〉.
Zaeh inge , J.G., Messe li, P., Gige , M., Ki eme, B., A umane, A., Da Sil a, M., Ecke , S.,
2021. La ge-scale ag icul u al in es men s in eas e n a ica: consequences o small-
scale a me s and he en i onmen . Ecosys . People 17 (1), 342–357.
Zhang, C., Fu, S., 2009. Allelopa hic e ec s o eucalyp us and he es ablishmen o mixed
s ands o eucalyp us and na i e species. Fo . Ecol. Manag. 258 (7), 1391–1396.
C. Chia ella e al.
