Poverty and food security impacts of sustainable intensification: Evidence from Ethiopia

Sa iye , O khan; As a o , Jacob; Zelle , Man ed
A icle — Published Ve sion
Po e y and ood secu i y impac s o sus ainable
in ensi ica ion: E idence om E hiopia
Food Secu i y
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Sa iye , O khan; As a o , Jacob; Zelle , Man ed (2025) : Po e y and ood
secu i y impac s o sus ainable in ensi ica ion: E idence om E hiopia, Food Secu i y, ISSN
1876-4525, Sp inge Ne he lands, Do d ech , Vol. 17, Iss. 2, pp. 405-420,
h ps://doi.o g/10.1007/s12571-025-01517-9
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/323637
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/
Vol.:(0123456789)
Food Secu i y (2025) 17:405–420
h ps://doi.o g/10.1007/s12571-025-01517-9
ORIGINAL PAPER
Po e y and ood secu i y impac s o sus ainable in ensi ica ion:
E idence omE hiopia
O khanSa iye 1 · JacobAs a o 1 · Man edZelle 1
Recei ed: 13 May 2024 / Accep ed: 10 Janua y 2025 / Published online: 12 Feb ua y 2025
© The Au ho (s) 2025
Abs ac
As sus ainable in ensi ica ion is a majo pa hway o imp o ing ag icul u al p oduc i i y and educing he en i onmen al
impac s o land use, he Go e nmen o E hiopia and in e na ional de elopmen o ganiza ions ha e been p omo ing se e al
p ac ices and echnologies o sus ainable in ensi ica ion. Using panel da a om 368 a ming households in E hiopia om
2014, 2016, and 2019, his s udy gauges he po e y and ood secu i y impac s o In eg a ed Soil Fe ili y Managemen
echnologies and hei combined use wi h conse a ion ag icul u e p ac ices, speci ically minimum illage and c op o a-
ion.We ind signi ican posi i e e ec s o ISFM adop ion in e ms o inc easing die a y di e si y and ood expendi u e and
educing ood insecu i y. In e ms o po e y, ISFM adop ion dec eases he p obabili y o being poo , he po e y gap, and
he se e i y o po e y. When combined wi h CA p ac ices, we ind ha he e ec s a e consis en ly la ge o a me s who
in eg a e ISFM and CA o all ood secu i y and po e y measu es. Ou indings s ongly sugges ha he adop ion o ISFM
echnologies has signi ican posi i e implica ions o po e y educ ion and imp o ed ood secu i y. These bene i s a e likely
o gain a conside able boos i ISFM echnologies a e applied oge he wi h CA p ac ices.
Keywo ds Technology adop ion· In eg a ed soil e ili y managemen · Conse a ion ag icul u e· Mul inomial
endogenous swi ching eg ession
1 In oduc ion
The numbe o people acing hunge has been g owing since
2014, especially in A ica which has wi nessed a ecen
sha p inc ease in all egions o he con inen (FAO e al.,
2023). In A ica, con inued popula ion g ow h has inc eased
ood demand and clima e a iabili y has u he s ained
a ming sys ems ha ha e long been pe o ming below
hei p oduc i e po en ial (The Mon pellie Panel, 2013).
Ag icul u e is he backbone o he economy in mos A ican
coun ies, pa icula ly hose in sub-Saha an A ica (SSA),
and has a s ong co ela ion wi h economic g ow h (Jayne &
Sanchez, 2021). Thus, inc easing ag icul u al p oduc i i y
is an o e a ching goal o educing hunge and sus aining
economic g ow h in SSA coun ies.
Since 2000, c op p oduc ion in SSA has expe ienced sig-
ni ican g ow h. Howe e , much o his g ow h o igina es
om he expansion o he a ea unde cul i a ion, which
is no a sound s a egy on en i onmen al and ecological
g ounds (Jayne & Sanchez, 2021). In ligh o his, na ional
go e nmen s, de elopmen o ganiza ions, and ounda ions
ha e been p omo ing a ious sus ainable in ensi ica ion (SI)
app oaches as a “new pa adigm o A ican ag icul u e” in
many SSA coun ies (Pe e sen & Snapp, 2015; The Mon -
pellie Panel, 2013). Al hough he e m “sus ainable in ensi-
ica ion” lacks a p ecise de ini ion (Pe e sen & Snapp, 2015),
i is o en linked o ag icul u al echnologies and p ac ices
ha achie e mo e p oduce om he exis ing a ea unde
cul i a ion, while educing he en i onmen al oo p in o
ag icul u al p oduc ion (God ay e al., 2010; P e y e al.,
2011). Besides SI, se e al o he app oaches seek o educe
en i onmen al ex e nali ies and imp o e p oduc i i y, e.g.,
conse a ion ag icul u e (CA), ag oecology, and ecologi-
cal in ensi ica ion. In e ms o SI, In eg a ed Soil Fe ili y
Managemen (ISFM) echnologies, which align well wi h
* O khan Sa iye
o.sa iye[email p o ec ed]
Jacob As a o
jacob.as a [email p o ec ed]
Man ed Zelle
[email p o ec ed]
1 Depa men o Ru al De elopmen Theo y andPolicy,
Uni e si y o Hohenheim, S u ga , Ge many
406 O.Sa iye e al.
he p inciples o SI, ha e been ecognized as means o
expand ag icul u al p oduc i i y and p e en soil deg ada-
ion (P e y e al., 2011; Vanlauwe e al., 2015). Likewise,
CA has ecei ed signi ican a en ion due o i s po en ial o
imp o e soil quali y. Ru al households a e likely o adop a
combina ion o a ious app oaches and p ac ices. This pape
seeks o answe he ques ion o how SI impac s po e y and
ood secu i y among u al smallholde s in E hiopia by u iliz-
ing h ee ounds o panel da a om E hiopia and concen a -
ing on ISFM echnologies and hei combina ion wi h CA.
Ag icul u al in ensi ica ion ia ISFM is conside ed he
leading pa hway o he A ican G een Re olu ion (Pe e sen
& Snapp, 2015). ISFM aims o enhance he ag onomic use
e iciency o inpu s. I is de ined as a se o soil e ili y
managemen p ac ices which include e ilize , imp o ed
ge mplasm, and o ganic inpu s, such as manu e, compos ,
o c op esidues adap ed o he local con ex (Vanlauwe
e al., 2010). CA is ano he SI app oach, which is based on
minimal soil dis u bance, c op esidue e en ion, and c op
o a ion o in e c opping (Gille e al., 2009). The combina-
ion o CA p ac ices is likely o imp o e soil quali y and
p o i abili y, assuming he su icien a ailabili y o neces-
sa y machine y (Johansen e al., 2012) and inpu s, such as
e ilize and he bicides (Gille e al., 2009). ISFM equi es
up on in es men in pu chased inpu s, such as e ilize
and imp o ed seeds. CA p ac ices a e labo -in ensi e; hus,
hei adop ion may inc ease labo cos s (Mon & Luu, 2020;
Vanlauwe e al., 2010). While ISFM adop e s a e likely o
expe ience highe yields in he sho e m, he bene i s o CA
usually occu in he long e m (Gille e al., 2009).
Conside ing he inc easing e o s o p omo e SI in SSA,
s udies ha e in es iga ed he implica ions o di e en SI
p ac ices on a ious li elihood indica o s in di e en coun-
y con ex s. Howe e , despi e exis ing s udies (Hö ne &
Wollni, 2021, 2022; Kassie e al., 2018; Khonje e al., 2018;
Maggio e al., 2022; Teklewold e al., 2013a, 2013b) ha
examine a ious CA o ISFM p ac ices and echnologies
o hei combina ions, igo ous s udies on he po e y and
ood secu i y implica ions o ISFM echnologies a e scan .
Among hese s udies, in a s udy om Uganda, Maggio e al.
(2022) ind ha bo h o ganic e ilize and maize-legume
in e c opping inc eases he o al alue o c op p oduc ion.
Using wo ounds o da a om E hiopia, Kassie e al. (2018)
show ha di e en combina ions o e ilize , imp o ed
maize a ie ies, and legume di e si ica ion inc ease maize
yield and income. In ano he s udy om E hiopia, Teklewold
e al. (2013b) epo ha maize a me s achie e he high-
es income when imp o ed maize a ie ies, maize-legume
o a ions, and minimum illage a e adop ed in combina ions
compa ed o in isola ion. Tes aye e al. (2021) u he ind
ha ce eal-legume in e c opping, educed illage, and hei
combina ion educe po e y headcoun , he po e y gap,
and he se e i y o po e y in E hiopia. Khonje e al. (2018)
show ha Zambian a me s achie e highe yield and income
when imp o ed maize a ie y and CA a e join ly adop ed.
Mo i a ed by he ac ha mos o he e idence on he
impac o ISFM ype p ac ices is limi ed o c op yield and
e enue, Hö ne and Wollni (2021) s udy ISFM implica ions
on household income, ood secu i y, and child educa ion.
They ind posi i e e ec s o ISFM on hese wel a e indica-
o s depending on he ag o-ecological zone. Using he same
da ase , Hö ne and Wollni (2022) show ha ISFM p ac ices
and hei combina ions imp o e land and labo p oduc i i y.
Besides Hö ne and Wollni (2021), we a e no awa e o any
o he s udy ha in es iga es he e ec s o ISFM adop ion on
po e y o ood secu i y ou comes. Employing h ee wa es
o panel da a om a echnology adop ion su ey in E hiopia,
his s udy con ibu es o he li e a u e by concen a ing on
ISFM p ac ices and analyzing hei adop ion implica ions on
po e y and ood secu i y ou comes and by examining he
join e ec s o ISFM and CA adop ion on hese ou comes.
The pape p oceeds as ollows: Sec .2 p esen s he da a,
a iables o in e es , and he me hodology employed o he
analyses; esul s a e p esen ed in Sec .3 and discussed in
Sec .4; Sec .5concludes he pape . We p o ide de ailed
desc ip ion o he s udy con ex in he appendix.
2 Da a andme hods
This sec ion i s p esen s he sampling s a egy and da a
u ilized o his s udy be o e desc ibing in de ail he po e y
and ood secu i y measu es and he s a egy o empi ical
analyses.
2.1 Sampling andda a
We employ household-le el da a collec ed om a an-
dom sample o a m households in 15 wo edas (dis ic s)
ac oss 10 zones in sou hwes e n E hiopia. Figu e1 shows
he su ey loca ions which ha e a ying clima ic and
ag o-ecological cha ac e is ics. The sample is de i ed
om a sub-sample o a na ionally ep esen a i e base-
line su ey conduc ed in 2012 by he In e na ional Food
Policy Resea ch Ins i u e (IFPRI) and he Ag icul u al
T ans o ma ion Agency (ATA) o E hiopia.1 Due o
adminis a i e and logis ical cons ain s, he su eyed
households we e andomly selec ed om he lis o su -
eyed a me s si ua ed wi hin a adius o app oxima ely
200km a ound he own o Hawassa. The su ey was
di e en om he na ionally ep esen a i e baseline con-
duc ed in 2012 and was speci ically designed o collec -
ing da a on smallholde a me s' isk managemen and
1 Fo mo e de ails on his sample, see (Mino & Sawye , 2013).
407
Po e y and ood secu i y impac s o sus ainable in ensi ica ion
inno a ion s a egies and hei impac s on po e y and
esilience. The a eas co e pa s o Sou he n Na ions,
Na ionali ies, and Peoples’ (SNNP) and O omia egions.
Th ee ollow-up su eys we e conduc ed in 2014, 2016,
and 2019. This s udy does no use he 2012 baseline by
IFPRI and ATA because some impo an da a collec ed
in he ollow-up su eys we e no p esen in he baseline
da a. The sample consis s o a balanced panel o 376 a m
households; howe e , he da a employed in his s udy co -
e s 368 households due o missing consump ion da a o
eigh households in 2016. Using he geog aphical coo -
dina es o he sampled households, we supplemen he
su ey da a wi h his o ical ain all da a ex ac ed om
he Clima e Haza ds G oup In aRed P ecipi a ion wi h
S a ion da a. This is a quasi-global ain all da ase ha
inco po a es 0.05° esolu ion sa elli e image y wi h in-
si u s a ion da a o c ea e g idded p ecipi a ion da a (Funk
e al., 2015). We ex ac annual ain all da a be ween 1981
and 2018 and calcula e s anda d de ia ions and his o ical
a e ages o he men ioned pe iod.
2.2 Po e y and ood secu i y measu es
Adop ion o SI echnologies and p ac ices is expec ed o p o-
mo e p oduc i i y and esilience among u al a me s. Highe
p oduc i i y and imp o ed esilience a e expec ed o con ib-
u e o educed po e y and imp o ed ood secu i y among
he adop e s o SI echnologies and p ac ices. In his s udy,
we employ di e en measu es o po e y and ood secu i y
which e lec sho - and long- e m ou comes. We assume ha
po e y ou comes acc ue o long- e m sus ained p oduc i i y
gains om SI. Food secu i y ou comes like ood expendi u e
and die a y di e si y a e likely o de i e om sho - e m p o-
duc i i y gains. Mo eo e , hese ou comes can also e lec he
long- e m esilience o he households.
Following Tes aye e al. (2021), we employ he Fos e -
G ee -Tho beck me hod in calcula ing he a ious po e y
indices (Fos e e al., 1984):
(1)
P
𝛼=1
N
N
∑
i=1
(z−yi
z)
𝛼
I(yi<z)
,
Fig. 1 Loca ion o sampled households
408 O.Sa iye e al.
whe e
yi
deno es he pe adul equi alen mon hly consump-
ion expendi u e de la ed by he consume p ice index using
2016 p ices as he base,
N
deno es he sample size,
z
deno es
he na ional po e y line,2 and
I(yi<z)
akes he alue o 1
when he
i h
household has a pe adul equi alen mon hly
consump ion expendi u e below he na ional po e y line.
Because we a e no in e es ed in he ac ion o he popula-
ion o sample below he po e y line, we emo e
1
N
om he
o mula. We calcula e h ee po e y indices: when
𝛼=0
,
P
is he po e y headcoun which indica es i he
i h
household
is poo ; when
𝛼=1
,
P
deno es he po e y gap index which
shows how a he
i h
household is om he po e y line; and
when
𝛼=2
,
P
indica es he squa ed po e y gap o po e y
se e i y index which e lec s inequali y among he poo .
We adop h ee dis inc indica o s o asce ain he ood
secu i y s a us o he sampled households. The household
die a y di e si y sco e (HDDS) is a commonly employed
measu e o household le el die a y quali y and ood secu i y.
HDDS desc ibes he household’s abili y o access di e -
en ood i ems. Ruel (2003) concludes ha conside ing i s
associa ion wi h ene gy a ailabili y and pe capi a income,
cap u ing die a y di e si y is a p ac ical me hod o unde -
s anding he ood secu i y s a us o households. The su ey
employed a se en-day ecall pe iod o cap u e day- o-day
a ia ion in household die s. Using his da a, we calcula e
HDDS based on 12 ood g oups: ce eals; whi e ube s and
oo s; ege ables; ui s; mea ; eggs; ish and o he sea ood;
legumes, nu s, and seeds; milk and milk p oduc s; oils and
a s; swee s; and spices, condimen s, and be e ages (Ken-
nedy e al., 2011). Ou second measu e o ood secu i y is
he p obabili y o expe iencing low ood a ailabili y in a
leas one mon h o e he pas 12mon hs. Finally, we cal-
cula e he adul equi alen eal ood expendi u e based on
he epo ed mon hly ood expendi u e3 de la ed by he con-
sume p ice index using 2016 p ices as he base.
2.3 Empi ical s a egy
Ou objec i e is o unpack he adop ion implica ions o
ISFM p ac ices and hei combina ion wi h CA on po -
e y and ood secu i y. The CA p ac ices conside ed in his
s udy a e minimum illage and c op o a ion. A household
is conside ed an adop e o CA p ac ices i ei he minimum
illage o c op o a ion o bo h a e adop ed. To unde s and
he impac s o echnologies and p ac ices adop ed by house-
holds, o each household we mus compa e he espec i e
ou comes om adop ion o he ou comes om non-adop ion.
Howe e , o each household, we can only obse e one s a e
a a ime because each household ei he adop s a ce ain
combina ion o ISFM p ac ices (o hei combina ion wi h
CA) o does no adop . Thus, we ha e a missing coun e ac-
ual which is he undamen al challenge o causal in e ence.
In his case, we need o compa e adop e s wi h non-adop e s,
which leads o ano he challenge – selec ion bias o endoge-
nei y (Imbens & Woold idge, 2009). Ag icul u al echnol-
ogy adop ion is a choice ou come which can be in luenced
by some unobse able ac o s, such as skills, mo i a ion,
ca e o amily, bond be ween amily membe s, and hei
dedica ion o a common goal, ha can also in luence ood
secu i y and/o po e y ou comes. A ailabili y o panel da a
sol es he p oblem i he unobse ed he e ogenei y is ime-
in a ian . Howe e , he bias s ill a ises i he unobse ed
he e ogenei y is ime- a ian .
To o e come po en ial endogenei y, we apply he mul i-
nomial endogenous swi ching eg ession (MESR) model,
which has been employed in se e al echnology adop ion
s udies (Bi u e al., 2020; Hö ne & Wollni, 2022; Kassie
e al., 2015, 2018; Khonje e al., 2018; Mon & Luu, 2020;
Teklewold e al., 2013b; Tes aye e al., 2021). MESR es i-
ma es sepa a e ou come eg essions o adop e s and non-
adop e s, which allows o an in e ac ion be ween he adop-
ion decision and obse ed and unobse ed he e ogenei y.
This ensu es ha adop ion has an impac on bo h he slope
and in e cep in he ou come equa ions. The es ima ed ou -
comes ep esen he e u ns o he cha ac e is ics o adop e s
and non-adop e s (Bi u e al., 2020; Di Falco e al., 2011;
Kassie e al., 2018).
Gi en ha we examine he impac s o ISFM adop ion
and he combina ion o ISFM and CA, we un wo sepa-
a e MESR.4 He e, we ake ISFM impac es ima ion as an
example o desc ibe he implemen a ion o he me hod.
MESR is implemen ed in wo s eps. In he i s s ep, we
model he de e minan s o ISFM adop ion. We assume ha
a each pe iod, ame s selec an ISFM echnology se ha
maximizes hei u ili y. Among he sampled households,
chemical e ilize is he dominan ISFM echnology ha
is adop ed. Hence, we conside chemical e ilize as a co e
componen and hypo hesize wo possible ISFM se s. One se
en ails he pa ial adop ion o wo ou o h ee co e ISFM
echnologies, speci ically, ei he chemical e ilize wi h an
2 The Cen al S a is ical Agency (2017) epo s ha he na ional
po e y line o 2015/2016 is 7,184 E hiopian Bi pe yea , which is
equal o 599 E hiopian Bi pe mon h pe adul pe son.
3 This measu e consis s o wo pa s: ood consumed a home which
can be pu chased o p oduced (based on he epo ed ma ke p ice),
and ood consumed away om home.
4 CA p ac ices only include conse a ion illage and c op o a ion.
C op esidue (i.e. mulch o compos ) is conside ed as an ISFM ech-
nology in his s udy. Fo mulch, c op o a ion, and conse a ion ill-
age, households a e conside ed as adop e s i hey apply he p ac ice
on a leas 10% o hei a mland. Please e e o he appendix o u -
he de ail on he echnology combina ions ha cons i u e a combina-
ion o ISFM and CA p ac ices.

409
Po e y and ood secu i y impac s o sus ainable in ensi ica ion
imp o ed a ie y o wi h an o ganic inpu .5 The second se
comp ises comple e ISFM adop ion, meaning ha all h ee
ISFM componen s a e adop ed: chemical e ilize , imp o ed
seeds, and o ganic inpu s.
Since he ISFM p inciple equi es he simul aneous use
o hese echnologies, e en i a household adop s one ISFM
echnology, we code ha household as a non-adop e . Thus,
assuming a echnology se
k=0,1, 2
, whe e
k=0
indica es
ha no ISFM se was adop ed, we can show a me s’ andom
u ili y a a ce ain ime as
Uk
. A a ional a me will adop
a pa icula echnology se gi en ha he expec ed u ili y is
highe han he al e na i es, i.e.,
Uk >Um ,m≠k
. Follow-
ing Kassie e al. (2018) unde he assump ion o independ-
ence o i ele an al e na i es (IIA),6 we es ima e he i s
s age wi h he mul inomial logi model. The p obabili y ha
a household
i
wi h cha ac e is ics
X
will adop ISFM se
k
is gi en as:
whe e
i
deno es he household,
k
is he echnology se ,
indica es yea , αk is a cons an e m o he echnology se
j,βk indica es ep esen s pa ame e s o be es ima ed, and
Xi
indica es a ious household socio-economic cha ac e is ics,
su ey yea , and ain all a iables. To bene i om he panel
s uc u e o he da ase , we es ima e he pooled selec ion
and ou come (i.e., second s ep) models using he Mundlak
app oach and include he means o all ime- a ying a iables
in o bo h equa ions. The Mundlak de ice helps o con ol
ime-in a ian unobse ed he e ogenei y as wi h ixed-
e ec s.
Hi
in Eq.2 indica es he means o he ime- a ying
explana o y a iables. In he second s age, ou come models
a e es ima ed o adop e s o each echnology se s and non-
adop e s sepa a ely:
whe e
Oi
indica es he ou come (i.e., ood secu i y o po -
e y indica o s) obse ed o household
i
a ime
,
Zi
indi-
ca es a ious household socio-economic cha ac e is ics and
(2)
P ob�
k
�
Xi ,Hi
�
=𝑒𝑥𝑝
�
𝛼k+Xi 𝛽k+Hi
�
∑
k
m=1
exp(𝛼
m
+X
i
𝛽
m
+H
i
)
,k=
0, 1, 2
(3)
⎧
⎪
⎨
⎪
⎩
Regime0∶Oi 0=Zi 0β0+

λi 0σ0+Hi0+ε
i 0i non −adop e
Regime1∶Oi 1=Zi 1β1+

λi 1σ1+Hi1+ε
i 1i pa ialISFMadop e
Regime2∶Oi 2=Zi 2β2+

λi 2σ2+Hi2+ε
i 2i comple eISFMadop e
,
su ey yea s,
β
deno es pa ame e s o be es ima ed,
Hi
indi-
ca es means o ime- a ying explana o y a iables, and
σ
ep esen s he co a iance be ween ou come and selec ion
equa ion e o s. An impo an e m in Eq.3 is

λ
, which
ep esen s he in e se Mills a ios de i ed om he selec ion
equa ion o cap u e ime- a ian indi idual e ec s. In e se
Mills a ios gene a ed om he i s s age help imp o e
iden i ica ion; howe e , i is impo an ha he i s s age
includes a selec ion ins umen .
We include wo ins umen al a iables in he i s s age:
he s anda d de ia ion o ain all be ween 1981 and 2018
and he his o ical a e age o he same pe iod. We assume
ha his o ical ain all pa e ns can play a ole in echnol-
ogy choices and adop ion decisions o a me s, bu ha hey
a e no di ec ly ela ed o he ou comes o in e es . How-
e e , he e is a possibili y ha ain all a ia ion could a ec
ou come a iables o in e es h ough non- a m income.
D ough (i.e., wi hin he las g owing season) leads o
inc eased employmen in o - a m sel -employmen bu no
in o - a m wage-employmen in E hiopia (Musungu e al.,
2024). Al hough he e idence p o ided is o he las g ow-
ing season, we can assume ha his o ical a ia ion can a ec
a me s’ expec a ions om ag icul u al p oduc ion. Con ol-
ling o non- a m income, we block he e ec o he ins u-
men s on ou come a iables possible ope a ing h ough non-
a m income and achie e condi ional independence o he
ins umen s.
Teklewold e al. (2013a) show ha he p obabili y o
manu e applica ion is high in a eas whe e ain all is eli-
able in e ms o dis ibu ion, ime, and amoun . De con
and Ch is iaensen (2011) ind ha e ilize applica ion is
p o i able o ce eals in E hiopia, gi en ha ain all does
no la gely de ia e om i s usual pa e ns. Fa me s do no
bene i om e ilize i ain all is no su icien . Kassie e al.
(2013) ind ha a me s in Tanzania choose o cul i a e a-
di ional a ie ies o e imp o ed a ie ies i he expec ed
ain all is insu icien . This implies ha in he absence o a
eliable ain all amoun , a me s do no in es in expensi e
inpu s. Conside ing hese esul s, we can assume ha ain-
all de ia ion and he his o ical a e age o ain all a e nega-
i ely and posi i ely associa ed wi h he adop ion o ISFM
p ac ices, espec i ely. In e ms o CA p ac ices, Teklewold
e al. (2013a) epo ha simila o manu e applica ion, he
p obabili y o adop ing c op o a ion is high in a eas whe e
ain all is eliable. A slan e al. (2014) ind a posi i e asso-
cia ion be ween dis ic -le el his o ical ain all a ia ion and
he adop ion and in ensi y o minimum illage. Gi en hese
indings, we again assume ha ain all a ia ion and he his-
o ical a e age o ain all a e de e minan s o he adop ion
o he CA p ac ices conside ed in his s udy.
Equa ion3 is used o p edic obse ed and coun e ac ual
ou comes o adop e s o he di e en echnology se s. The
5 We conside manu e, mulch, and compos as o ganic inpu s unde
ISFM. Un o una ely, he da a does no allow us o di e en ia e
o ganic mulch ma e ials om o he mulch ma e ials, such as plas ics.
6 Bou guignon e al. (2007) show ha selec ion bias co ec ion based
on he mul inomial logi model p o ides su icien co ec ion o he
second s age ou come equa ion e en i IIA is iola ed. In his s udy,
we employed Dubin and Mc Fadden’s Model a ian 1 using he
“selmlog” S a a command de eloped by Bou guignon e  al. (2007)
wi h 200 boo s apped eplica ions.
410 O.Sa iye e al.
obse ed ou comes a e compu ed gi en he in o ma ion in
he da a as:
whe e
Zi
deno es obse ed means o ime- a ying explana-
o y a iables; β,σ, and θ a e he coe icien s es ima ed o
adop e s. To be able o es ima e a e age ea men e ec s on
he ea ed, i is impo an o also calcula e he coun e ac-
ual ou comes o adop e s o echnology se s. The coun e -
ac uals o adop e s a e calcula ed as ollows:
Pa ame e s in Eq.5 a e coe icien s ob ained om he
ou come eg essions o non-adop e s. Wi h his equa ion,
we calcula e he ou comes ha adop e s would ha e ob ained
i he e u ns o hei cha ac e is ics had been he same as he
e u ns obse ed by non-adop e s. The s anda d e o s a e
ob ained by 200 boo s apped eplica ions o accoun o he
i s s age. The di e ence be ween Eqs.4 and 5 p o ides us
wi h he a e age ea men e ec s on he ea ed:
The e m
(βk−β
0)Zi k
ep esen s he change obse ed
in he ou come due o di e ences in obse ed cha ac e -
is ics. The second and hi d e ms, namely
(σk
−σ
0
)

λ
i k
and
(
𝜃
k
−𝜃
0
)Z
ik
, deno e he di e ences obse ed due o
ime- a ian and ime-in a ian unobse ed cha ac e is ics,
espec i ely. Conside ing ha we es mul iple hypo heses,
he e is a p obabili y o ejec ing a ue null hypo hesis.
Hence, we con ol o he alse disco e y a e and calcula e
and epo he sha pened q- alues as sugges ed by Ande son
(2008).
3 Resul s
This sec ion p o ides an o e iew o he su eyed house-
holds be o e p esen ing he esul s o he impac s o SI on
ood secu i y and po e y ou comes.
3.1 Sample cha ac e is ics
This subsec ion epo s da a on impo an sample cha ac-
e is ics. Table1 lis s ele an desc ip i e s a is ics, and
Fig.2 depic s adop ion ends o ISFM and CA p ac ices
and echnologies. Among he ou come a iables, HDDS was
close o six ood g oups in 2014, while in he la e ounds, i
inc eased o se en ood g oups, indica ing an imp o emen
in ood secu i y. Al hough eal pe capi a ood expendi u e
has declined o e he yea s, he p obabili y o expe iencing
(4)
E
(O
i K |
k=K)=Z
i k
β
k
+

λ
i k
σ
k
+Z
ik
𝜃
k
,k=
1,2
(5)
E(
O
i 0
|k
=
K
)=
Z
i kβ0+
λi kσ0+
Z
ik
𝜃
0
,k
=1,2
(6)
ATT
i
=E(O
i K
|k=K)−E(O
i 0
|k=K)=
(βk
−β
0
)Z
i k
+ (σ
k
−σ
0
)

λ
i k
+(𝜃
k
−𝜃
0
)Z
ik.
ood insecu i y has also declined signi ican ly om he i s
su ey ound o he la e ounds. In e ms o po e y indi-
ca o s, all show ha po e y and i s se e i y a e g owing
among he sampled households. Rega ding household cha -
ac e is ics, he sampled households ha e, on a e age, six
membe s and 17% a e headed by women. I appea s ha
households ha e ex ended hei social ne wo ks by aking
ac i e oles in social o ganiza ions. App oxima ely, 16% o
households ecei ed loans and a ound 34% had non- a m
income in all su ey ounds.
Be ween 2014 and 2016, he a e age a m size inc eased,
bu did no change in 2019. The p ima y c ops a e ba ley and
maize, wi h he sha e o a me s cul i a ing maize inc easing
o 40% in 2019, om app oxima ely 33% in he p e ious su -
ey ounds. Addi ionally, app oxima ely 42% o he sampled
households cul i a e ba ley. T opical li es ock uni s (TLU)
owned inc eased be ween 2014 and 2016, bu declined in
2019. We obse e g ow h in he numbe o a me s who
pe cei e hei soil quali y o be ei he poo o good. O e
he yea s, he e ha e been imp o emen s in in as uc u e, as
indica ed by less ime needed o each he nea es pe iodic
ma ke and ex ension o ice. The sha e o a me s a end-
ing ex ension aining has inc eased o e he yea s. How-
e e , we obse e a decline in he sha e o a me s ecei ing
ex ension isi s in 2019. On he o he hand, he sha e o
households e e ing o hei ellow a me s o in o ma ion
ega ding ag icul u al p ac ices and echnologies inc eased
signi ican ly, especially in 2019. In e ms o ain all, he yea
2015 was, on a e age, d ie han 2013 and 2018.
We obse e ha he numbe o ISFM adop e s7 inc eased
om one su ey ound o he nex . In e ms o ISFM and
CA combina ion, he numbe o a me s implemen ing
bo h p ac ices inc eased om 2014 o 2016, and in 2019
i emained close o he 2016 alues. O e all, he numbe
o a me s who do no p ac ice any combina ion o ISFM
o CA p ac ices declined signi ican ly om 2014 o 2016.
The numbe o ISFM adop e s who do no p ac ice CA also
inc eased om one su ey ound o he nex . Mo eo e , he
numbe o a me s who p ac ice only CA inc eased om
2014 o 2016, bu e u ned o 2014 le els in 2019.
Among he ISFM echnologies, e ilize applica ion was
he dominan componen , wi h an inc eased sha e o house-
holds using e ilize o e he yea s. In 2019, almos 80% o
he sampled households indica ed applying some e ilize .
The sha e was 72% in 2014 and 77% in 2016. Rega ding
manu e, 23%, 44%, and 49% o he sampled households
epo ed i s use in 2014, 2016, and 2019, espec i ely.
Households ha e also inc eased hei use o imp o ed a ie-
ies o e he yea s. Mulch was u ilized by 9%, 6%, and 8% o
7 Those who use chemical e ilize wi h ei he manu e o imp o ed
a ie y (pa ial ISFM) o bo h (comple e ISFM).
411
Po e y and ood secu i y impac s o sus ainable in ensi ica ion
Table 1 Sample cha ac e is ics
No es: ISFM is In eg a ed Soil Fe ili y Managemen ; CA is conse a ion ag icul u e; HDDS is he house-
hold die a y di e si y sco e; HH is household; and TLU is opical li es ock uni s. a dis ance, soil quali y,
and land ce i ica e a e o he bigges pa cel; soil quali y is based on a me s’ pe cep ions; b pa ial adop-
ion o ISFM: ei he chemical e ilize wi h imp o ed a ie y o wi h o ganic inpu and comple e adop-
ion o ISFM: all echnologies, chemical e ilize , imp o ed seeds, and o ganic inpu ; c Pa ial o comple e
ISFM wi h c op o a ion and/o minimum illage as CA;
2014 2016 2019
Household cha ac e is ics
HDDS 6.36 (1.68) 6.71 (1.51) 6.67 (1.63)
Adul equi alen eal ood expendi u e (ln) 6.25 (0.73) 6.18 (0.66) 6.09 (0.77)
Food insecu i y expe ience (= 1 i yes) 0.67 (0.47) 0.39 (0.49) 0.39 (0.49)
Po e y headcoun (= 1 i poo ) 0.60 (0.49) 0.62 (0.49) 0.65 (0.48)
Po e y gap 0.22 (0.25) 0.22 (0.22) 0.26 (0.26)
Po e y se e i y 0.11 (0.16) 0.10 (0.14) 0.13 (0.18)
Female head 0.17 (0.37) 0.16 (0.37) 0.17 (0.38)
Head age 44.90 (13.85) 48.15 (13.41) 49.71 (14.53)
Yea s o o mal schooling o head 3.17 (3.61) 3.12 (3.58) 3.13 (3.35)
Numbe o household membe s 6.37 (2.24) 6.45 (2.33) 6.33 (2.30)
Dependency a io 1.20 (0.88) 1.16 (0.84) 1.09 (0.86)
Fa m-speci ic cha ac e is ics
Dis ance o pe iodic ma ke (minu es) 52.63 (45.18) 47.84 (40.24) 40.90 (34.37)
Dis ance o ex . o ice (minu es) 46.46 (46.04) 37.53 (39.04) 31.25 (31.90)
Dis ance o pa cel (minu es) a14.90 (22.31) 20.48 (34.65) 18.78 (32.62)
Land ce i ica e (= 1 i yes) a0.83 (0.37) 0.67 (0.47) 0.78 (0.42)
Poo soil quali y (= 1 i yes) a0.04 (0.20) 0.03 (0.17) 0.08 (0.27)
Good soil quali y (= 1 i yes) a0.52 (0.50) 0.52 (0.50) 0.64 (0.48)
Fa m size in hec a es 1.47 (1.20) 1.67 (1.56) 1.67 (2.52)
TLU owned 3.88 (4.32) 4.59 (4.64) 3.96 (4.10)
Pes icides (= 1 i applied) 0.34 (0.47) 0.30 (0.46) 0.42 (0.49)
Maize (= 1 i p oduced in p e . season) 0.34 (0.48) 0.32 (0.47) 0.40 (0.49)
Ba ley (= 1 i p oduced in p e . season) 0.42 (0.49) 0.43 (0.50) 0.41 (0.49)
Numbe o shocks 0.36 (0.58) 0.25 (0.51) 0.42 (0.61)
No. o social o ganiza ions HH is ac i e in 1.41 (1.06) 1.84 (1.45) 2.13 (1.52)
Loan (= 1 i acqui ed in pas 12mon hs) 0.15 (0.35) 0.16 (0.37) 0.17 (0.37)
Non- a m income (= 1 i yes) 0.34 (0.47) 0.33 (0.47) 0.36 (0.48)
Ex ension isi s (= 1 i yes in pas 24mon hs) 0.39 (0.49) 0.39 (0.49) 0.17 (0.37)
Ex . aining (= 1 i yes in pas 24mon hs) 0.44 (0.50) 0.43 (0.50) 0.54 (0.50)
Fellow a me as in o ma ion sou ce (= 1 i yes) 0.21 (0.41) 0.30 (0.46) 0.63 (0.48)
Annual ain all p e ious season (mm) 1286.47 (301.01) 1095.16 (269.27) 1328.32 (263.97)
ISFM b (% in pa en heses)
No ISFM combina ion 207 (56.3%) 162 (44.0%) 142 (38.6%)
Pa ial adop ion o ISFM echnologies 115 (31.3%) 122 (33.2%) 143 (38.9%)
Comple e adop ion o ISFM echnologies 46 (12.5%) 84 (22.8%) 83 (22.6%)
ISFM and CA combina ion c
No ISFM combina ion o CA 136 (37.0%) 77 (20.9%) 75 (20.4%)
ISFM + CA 111 (30.2%) 135 (36.7%) 138 (37.5%)
CA 71 (19.3%) 85 (23.1%) 67 (18.2%)
ISFM 50 (13.6%) 71 (19.3%) 88 (23.9%)
Obse a ions 368 368 368
412 O.Sa iye e al.
he sampled households in he yea s 2014, 2016, and 2019,
espec i ely. C op o a ion was p ac iced by 48%, 59%, and
55% o he sampled households in 2014, 2016, and 2019,
espec i ely. The e was a signi ican g ow h in he sha e o
a me s p ac icing compos ing o e he yea s, i.e., 22% in
2014, 35% in 2016, and 42% in 2019. Ve y ew households
epo ed educed illage up ake.
3.2 Impac s o SI
The objec i e o his s udy is o e alua e he impac s o
widely p omo ed SI p ac ices on po e y and ood secu i y
among u al smallholde a me s in E hiopia. In his subsec-
ion, we concen a e on ISFM adop ion and i s combined
use wi h CA be o e p esen ing hei e ec s on ou come
measu es.
Table2 p esen s he mul inomial logis ic eg ession
esul s on ISFM adop ion and he adop ion o he combina-
ion o ISFM and CA. We pe o m he Hausman es o IIA
and do no ind any e idence o iola ing he IIA assump-
ion. This implies ha he mul inomial logis ic eg ession
model can be applied o ou da a. We obse e a signi ican
posi i e e ec o pes icide applica ion on he adop ion o
ISFM echnologies and CA p ac ices, and hei combined
use. Maize and ba ley p oduce s a e likely o adop ISFM
echnologies and hei combina ion wi h CA p ac ices. We
ind ha a me s who a e loca ed in high al i ude a eas a e
less likely o adop hese p ac ices. The esul s u he show
ha pa icipa ing in ex ension aining p og ams and e e -
ing o a ellow a me o ag icul u al- ela ed in o ma ion
a e posi i ely associa ed wi h he up ake o he combina ion
o ISFM and CA, as well as he adop ion o CA alone.
Rega ding he ins umen s, as hypo hesized, he esul s
sugges ha his o ical a e age ain all has a posi i e link
wi h ISFM adop ion in bo h speci ica ions. The s anda d
de ia ion o his o ical ain all is posi i ely co ela ed wi h
he adop ion o CA p ac ices and he combined use o ISFM
and CA. To de e mine whe he he ins umen s a e admis-
sible, we ollow Di Falco e al. (2011) and pe o m a simple
alsi ica ion es whe e he ins umen s a e assumed o a ec
adop ion decisions, bu no he ou come o non-adop e s.
TablesA1 and A2 in he appendix epo he esul s om
his simple alsi ica ion es . The esul s do no p o ide any
consis en e idence o a signi ican ela ionship be ween he
ins umen s and he ou comes o non-adop e s a 5% e o
p obabili y. This sugges s ha he ins umen s a e alid:
hey a e join ly signi ican de e minan s o adop ion deci-
sion bu no he ou comes o households who do no adop
echnology se s unde conside a ion. Mo eo e , we un an-
dom e ec s, ixed e ec s, and In e se P obabili y Weigh ed
Reg ession Adjus men (sepa a ed o each ound) impac
es ima ion echniques o assess he obus ness o he esul s.
TablesA5 -A9 depic he esul s om hese al e na i e ech-
niques. Mos o he esul s a e in line wi h he MESR esul s.
To isualize and compa e he magni ude o he impac s,
we depic he impac s o pa ial and comple e ISFM adop-
ion s. no ISFM adop ion in Fig.3 and Fig.4, espec-
i ely. These esul s a e also a ailable in able o ma in
he supplemen a y appendix (in TablesA3 and A4). The
igu es illus a e he obse ed and coun e ac ual ou comes
and he di e ence be ween hem. The e a e posi i e a e -
age ea men e ec s o pa ial and comple e adop e s o
ISFM echnologies in e ms o HDDS and adul equi alen
pe capi a ood expendi u e. Also, bo h pa ial and com-
ple e adop ion o ISFM echnologies educes he p obabil-
i y o expe iencing ood insecu i y o e he pas 12mon hs
by 8% and 6%, espec i ely. Bo h pa ial and comple e
adop e s o ISFM echnologies would ha e a ound 13%
Fig. 2 Adop ion o ISFM and
CA p ac ices, by yea
419
Po e y and ood secu i y impac s o sus ainable in ensi ica ion
Scien i ic Da a, 2(1), 150066. h ps:// doi. o g/ 10. 1038/ sda a. 2015.
66
Gille , K. E., Wi e , E., Co beels, M., & Ti onell, P. (2009). Conse a-
ion ag icul u e and smallholde a ming in A ica: The he e ics’
iew. Field C ops Resea ch, 114(1), 23–34. h ps:// doi. o g/ 10.
1016/j. c . 2009. 06. 017
God ay, H. C. J., Bedding on, J. R., C u e, I. R., Haddad, L., Law-
ence, D., Mui , J. F., P e y, J., Robinson, S., Thomas, S. M.,
& Toulmin, C. (2010). Food secu i y: The challenge o eeding
9 billion people. Science, 327(5967), 812–818. h ps:// doi. o g/
10. 1126/ scien ce. 11853 83
Hö ne , D., & Wollni, M. (2021). In eg a ed soil e ili y manage-
men and household wel a e in E hiopia. Food Policy, 100,
102022. h ps:// doi. o g/ 10. 1016/j. oodp ol. 2020. 102022
Hö ne , D., & Wollni, M. (2022). Does in eg a ed soil e ili y man-
agemen inc ease e u ns o land and labo ? Ag icul u al Eco-
nomics, 53(3), 337–355. h ps:// doi. o g/ 10. 1111/ agec. 12699
Imbens, G. W., & Woold idge, J. M. (2009). Recen de elopmen s in
he econome ics o p og am e alua ion. Jou nal o Economic
Li e a u e, 47(1), 5–86. h ps:// doi. o g/ 10. 1257/ jel. 47.1.5
Jayne, T. S., & Sanchez, P. A. (2021). Ag icul u al p oduc i i y
mus imp o e in Sub-Saha an A ica. Science, 372(6546),
1045–1047. h ps:// doi. o g/ 10. 1126/ scien ce. ab 54 13
Johansen, C., Haque, M. E., Bell, R. W., Thie elde , C., & Esdaile,
R. J. (2012). Conse a ion ag icul u e o small holde ain-
ed a ming: Oppo uni ies and cons ain s o new mechanized
seeding sys ems. Field C ops Resea ch, 132, 18–32. h ps:// doi.
o g/ 10. 1016/j. c . 2011. 11. 026
Kassie, M., Jale a, M., Shi e aw, B., Mmbando, F., & Meku ia, M.
(2013). Adop ion o in e ela ed sus ainable ag icul u al p ac-
ices in smallholde sys ems: E idence om u al Tanzania.
Technological Fo ecas ing and Social Change, 80(3), 525–540.
h ps:// doi. o g/ 10. 1016/j. ech o e. 2012. 08. 007
Kassie, M., Teklewold, H., Jale a, M., Ma enya, P., & E ens ein, O.
(2015). Unde s anding he adop ion o a po olio o sus ainable
in ensi ica ion p ac ices in eas e n and sou he n A ica. Land
Use Policy, 42, 400–411. h ps:// doi. o g/ 10. 1016/j. landu sepol.
2014. 08. 016
Kassie, M., Ma enya, P., Tessema, Y., Jale a, M., Zeng, Di., E en-
s ein, O., & Rahu , D. (2018). Measu ing a m and ma ke le el
economic impac s o imp o ed maize p oduc ion echnologies in
E hiopia: E idence om panel da a. Jou nal o Ag icul u al Eco-
nomics, 69(1), 76–95. h ps:// doi. o g/ 10. 1111/ 1477- 9552. 12221
Kennedy, G., Balla d, T., & Dop, M. C. (2011). Guidelines o measu -
ing household and indi idual die a y di e si y. Food and Ag icul-
u e O ganiza ion o he Uni ed Na ions.
Khonje, M. G., Manda, J., Mkandawi e, P., Tu a, A. H., & Alene, A.
D. (2018). Adop ion and wel a e impac s o mul iple ag icul u al
echnologies: E idence om eas e n Zambia. Ag icul u al Eco-
nomics, 49(5), 599–609. h ps:// doi. o g/ 10. 1111/ agec. 12445
Maggio, G., Mas o illo, M., & Si ko, N. J. (2022). Adap ing o high
empe a u es: E ec o a m p ac ices and hei adop ion du a ion
on o al alue o c op p oduc ion in Uganda. Ame ican Jou nal
o Ag icul u al Economics, 104(1), 385–403. h ps:// doi. o g/ 10.
1111/ ajae. 12229
Mino , N., & Sawye , B. (2013). Ag icul u al p oduc ion in E hiopia:
Resul s o he 2012 ATA baseline su ey. h ps:// eap. i p i. in o/
iles/ 2013/ 12/ ag_ p odu c ion_ in_ e hio pia. pd
Mon , G., & Luu, T. (2020). Does conse a ion ag icul u e change
labou equi emen s? E idence o sus ainable in ensi ica ion in
Sub-Saha an A ica. Jou nal o Ag icul u al Economics, 71(2),
556–580. h ps:// doi. o g/ 10. 1111/ 1477- 9552. 12353
Musungu, A. L., Kubik, Z., & Qaim, M. (2024). D ough shocks and
labou ealloca ion in u al A ica: E idence om E hiopia. Eu o-
pean Re iew o Ag icul u al Economics, jbae020. h ps:// doi. o g/
10. 1093/ e ae/ jbae0 20
Pe e sen, B., & Snapp, S. (2015). Wha is sus ainable in ensi ica ion?
Views om expe s. Land Use Policy, 46, 1–10. h ps:// doi. o g/
10. 1016/j. landu sepol. 2015. 02. 002
P e y, J., Toulmin, C., & Williams, S. (2011). Sus ainable in ensi ica-
ion in A ican ag icul u e. In e na ional Jou nal o Ag icul u al
Sus ainabili y, 9(1), 5–24. h ps:// doi. o g/ 10. 3763/ ijas. 2010. 0583
Ruel, M. T. (2003). Ope a ionalizing die a y di e si y: A e iew o
measu emen issues and esea ch p io i ies. The Jou nal o Nu i-
ion, 133(11 Suppl 2), 3911S-3926S. h ps:// doi. o g/ 10. 1093/ jn/
133. 11. 3911S
Tambo, J. A., & Mockshell, J. (2018). Di e en ial impac s o conse -
a ion ag icul u e echnology op ions on household income in
Sub-Saha an A ica. Ecological Economics, 151, 95–105. h ps://
doi. o g/ 10. 1016/j. ecole con. 2018. 05. 005
Teklewold, H., & Mekonnen, A. (2017). The illing o land in a chang-
ing clima e: Empi ical e idence om he Nile Basin o E hiopia.
Land Use Policy, 67, 449–459. h ps:// doi. o g/ 10. 1016/j. landu
sepol. 2017. 06. 010
Teklewold, H., Kassie, M., & Shi e aw, B. (2013a). Adop ion o mul i-
ple sus ainable ag icul u al p ac ices in u al E hiopia. Jou nal o
Ag icul u al Economics, 64(3), 597–623. h ps:// doi. o g/ 10. 1111/
1477- 9552. 12011
Teklewold, H., Kassie, M., Shi e aw, B., & Köhlin, G. (2013b). C op-
ping sys em di e si ica ion, conse a ion illage and mode n seed
adop ion in E hiopia: Impac s on household income, ag ochemi-
cal use and demand o labo . Ecological Economics, 93, 85–93.
h ps:// doi. o g/ 10. 1016/j. ecole con. 2013. 05. 002
Tes aye, W., Blalock, G., & Ti i ayi, N. (2021). Clima e-Sma inno-
a ions and u al po e y in E hiopia: Explo ing impac s and
pa hways. Ame ican Jou nal o Ag icul u al Economics, 103(3),
878–899. h ps:// doi. o g/ 10. 1111/ ajae. 12161
The Mon pellie Panel. (2013). Sus ainable in ensi ica ion: A new
pa adigm o A ican ag icul u e.
Vanlauwe, B., Ba iono, A., Chianu, J., Gille , K. E., Me ckx, R., Mok-
wunye, U., Ohiokpehai, O., Pype s, P., Tabo, R., Shephe d, K. D.,
Smaling, E. M. A., Woome , P. L., & Sanginga, N. (2010). In e-
g a ed soil e ili y managemen . Ou look on Ag icul u e, 39(1),
17–24. h ps:// doi. o g/ 10. 5367/ 00000 00107 91169 998
Vanlauwe, B., Descheemaeke , K., Gille , K. E., Huising, J., Me ckx,
R., Nziguheba, G., Wend , J., & Zingo e, S. (2015). In eg a ed
soil e ili y managemen in Sub-Saha an A ica: Un a elling
local adap a ion. The Soil, 1(1), 491–508. h ps:// doi. o g/ 10. 5194/
soil-1- 491- 2015
Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o
ju isdic ional claims in published maps and ins i u ional a ilia ions.
O khan Sa iye is a pos doc o al
esea che a he Ins i u e o
Ag icul u al Sciences in he
T opics (Hans-Ru henbe g-Ins i-
u e), Uni e si y o Hohenheim,
Ge many. His p ima y in e es s
e ol e a ound ag icul u al and
de elopmen economics, wi h a
pa icula ocus on ood secu i y,
sus ainable de elopmen , and
women's empowe men . He is
especially in igued by u al
de elopmen ela ed policy and
p og am e alua ions, seeking o
e alua e hei e ec i eness and
iden i y a eas o imp o emen .

420 O.Sa iye e al.
Jacob As a o is a doc o al can-
dida e a he Ins i u e o Ag icul-
u al Sciences in he T opics
(Hans-Ru henbe g-Ins i u e), Uni-
e si y o Hohenheim, Ge many.
His esea ch in e es encompasses
issues o ag icul u al p oduc i i y
and e iciency analysis, ood secu-
i y, echnology and inno a ion
adop ion, and po e y educ ion.
He has keen in e es in issues
ela ing o how he adop ion o
ag icul u al echnologies and
inno a ions ansla e in o g ea e
ood secu i y and educed po e y
o u al a m amilies.
Man ed Zelle is a P o esso and
Chai o Ru al De elopmen
Theo y and Policy a he Hans-
Ru henbe g-Ins i u e, Uni e si y
o Hohenheim, Ge many. He
se ed as he Di ec o o he
Food Secu i y Cen e (2009–
2014), wo ked as a Senio
Resea ch Fellow and P og am
Leade o he In e na ional Food
Policy Resea ch Ins i u e (IFPRI)
om 1993 o 1999 and again
om 2014 o 2016, and held a
p o esso ship a he Uni e si y o
Goe ingen (1999–2005). His
academic wo k emphasizes
applied esea ch on he impac o ood, ag icul u e, and u al de elop-
men policies on income, po e y s a us, and ood secu i y. His aca-
demic jou ney includes s udies in Ag icul u al Sciences a he Uni e -
si y o Bonn, whe e he ea ned a doc o al i le in Ag icul u al
Economics in 1990.