The impact of farm input subsidies on women and children : evidence from Malawi

Mwale, Martin Limbikani (2022-04)

Thesis (PhD)--Stellenbosch University, 2022.

Thesis

ENGLISH SUMMARY: Women and children remain the most vulnerable demographic groups in Sub-Saharan Africa (SSA). However, Farm Input Subsidy Programs (FISP) that have been shown to improve household welfare in SSA, could be expected to change the situation of women and children, because they are part of poor households whose welfare is targeted. While evaluations of FISP focus on household-level effects, the well-being of the most vulnerable people inside these households - women and children, is not given adequate attention. Moreover, the traditional and household contexts in which the impacts of the subsidies manifest is not given much consideration. In this thesis, I attempt to address this gap using data from Malawi, a country that has one of the highest levels of gender inequality within SSA. In Malawi, a large-scale farm subsidy program registered gains in total maize production, and opposing kinship traditions that allow for different levels of women’s empowerment, co exist. I took advantage of a rich set of micro-level data that Malawi collects. These data are nationally representative and capture the situation of household members, including women and children. The main data set used were Malawi’s Living Standards Measurement Surveys-the Integrated Household Surveys (IHS). The IHS dis-aggregates data at community, household, and individual levels. I therefore used the IHS to capture the outcomes of women and children. Further, the IHS provides information on whether a household benefits from FISP or not, which is the main treatment of interest in this thesis. There are also other additional data that I used in specific chapters, to supplement the IHS, in answering my research questions. For instance, I linked the IHS with satellite data to verify the impact of FISP on crop yields. In addition, I used the Service Provision Assessment (SPA) survey, that captures information about health facilities in communities, to understand how FISP combines with maternal care, of good quality, to bolster child nutrition. I specifically used these data to understand whether women exert command over income from the FISP and whether children’s outcomes improve due to the FISP, in four broad objectives. In the first objective of the thesis, I re-examined the impact of FISP on maize productivity, to provide the context for the three other objectives that followed in the thesis. Unlike previous evidence which used survey-enumerated yields that are prone to Non-Classical Measurement Errors, I took a step further to eliminate strategic bias in reporting yields by using satellite-generated data. I found that FISP associated with increased maize yields only in the south of Malawi-a region characterised by highest poverty levels, less arable land, and high population density, but not in the north or centre-regions with opposite socioeconomic standing. I argue that the differences occurred, because in the south, the FISP targeted the poor-households that needed it the most to afford inputs. In the second chapter I examined whether long-term child nutrition, height-for-age, improved among children under the age of five in households that received the FISP. Building on previous evaluations which find that the FISP only improved short-term and medium-term nutrition, I extended the analysis to understand the heterogeneous impact of the FISP on a long-term nutritional outcome for children, height-for-age. The results revealed that the FISP improves height-for-age for children under the age of five, only when their mothers were exposed to Focused Antenatal Care (FANC) which provides advice on good diets for pregnant mothers and their children. It is likely that the FISP increased food availability, and provided liquidity that allowed the purchase of more nutritious food, leveraging advice from FANC. I further show that these, in utero nutrition investments sustained beyond the age of three only when the child was exposed to follow-up participation in a community nutrition program that caters for children under the age of five. In the third chapter of the thesis, I examined whether FISP allowed women to have agency over cash that is earned from maize sales. I added another contextual factor which is key to determining how women exercise their decision-making power in Malawi - post-marriage settlement practices within kinship traditions. My results revealed that the FISP does not improve women’s decision-making power over earnings from maize sales. In fact, the program shifts the decision-making power on earnings from women to their husbands in matrilocal communities. In matrilocal settings, married couples reside in the natal community of women, and the women own household land that is controlled by their maternal uncle. I showed that the gain in matrilocal husbands’ decision-making due to FISP is larger than the loss in women’s decision-making, which leads to the suspicion that the additional gain for men comes from the woman’s maternal uncle. Because men receive FISP on behalf of their households, the results suggest that husbands in matrilocal communities use the program to counterbalance their dis-advantage of not owning land. In the fourth chapter I examined the impacts of FISP on children’s school drop out. I find that FISP reduced matrilocal girls’ school drop out. The effects are largely driven by increased expenditure on girls’ education due to the program. Matrilocal parents likely invested some of the proceeds from FISP in girls, because after marriage, matrilocal women co reside with their parents, thus taking care of the parents when they age - a duty undertaken by boys in patrilocal communities. The investment in girls for future benefits of parents, is more relevant for matrilocal husbands because these husbands do not own land. Thus, even without controlling land, the husbands can at least receive care from their daughters. Further, I found that FISP does not affect school drop out in patrilocal communities. From a Malawian context of overall high female school drop out, this result is likely because boys, who are heirs in patrilocal communities, already have lower rates of drop out. I conclude that FISP associates with improved maize productivity when targeted at poor people, and FISP benefits younger children with increased nutrition. Further, the FISP reduces gender inequality only among school-going children in matrilocal communities, while it increases gender inequality by reducing decision-making among older women in the same communities. Alternative policies, that have proven key to improving women’s agency in other countries, such as direct cash transfers to women, should be promoted, because increasing agency through directly targeting the FISP at women is unlikely to assist. This is because de facto controllers of land, and its proceeds, are men-husbands in patrilocal communities, and maternal uncles and husbands in matrilocal communities. In addition, general welfare policy like FISP must account for traditional context for women and children, to avoid unintended negative consequences, such as worsening gender inequalities. Furthermore, even welfare strategies that are specific to improving the well-being of women and children, may need to be tailored to contextual factors such as kinship traditions, to yield wide-spread positive impacts on women’s and children’s well-being.

AFRIKAANSE OPSOMMING: Vroue en kinders bly die kwesbaarste demografiese groepe in Afrika Suid van die Sahara (SSA). Daar kan verwag word dat die “Farm Input Subsidy Programs” (FISP) die welstand van huishoudings in SSA sal verbeter, en die situasie van vroue en kinders sal verander omdat hulle deel uitmaak van arm huishoudings wie se welsyn geteiken word. Terwyl die evaluering van FISP word gefokus op gevolge op huishoudelike vlak, word die welstand van die kwesbaarste mense binne hierdie huishoudings, vroue en kinders, nie voldoende aandag gegee nie. Daar word nie veel aandag gegee aan die tradisionele en huishoudelike kontekste waarin die gevolge van die subsidies manifesteer nie. In hierdie tesis probeer ek om hierdie leemte aan te spreek met behulp van data vanuit Malawi, ’n land met een van die hoogste vlakke van geslags ongelykheid binne SSA. In Malawi het ’n grootskaalse boerdery subsidie program ’n toename in die totale mielie produksie en teenstrydige verwantskaps tradisies geregistreer wat verskillende vlakke van vroue bemagtiging moontlik maak. Ek het gebruik gemaak van ’n ryk stel data op mikrovlak wat Malawi versamel. Hierdie data is nasionaal verteenwoordigend en beskryf die situasie van huishoudelike lede, insluitend vroue en kinders. Die belangrikste datastel wat gebruik is, was Malawi se “Living Standards Measurement Surveys”, die “Integrated Household Surveys” (IHS). Die IHS verdeel data op gemeenskaps, huishoudelike en individuele vlak. Ek het dus die IHS gebruik om die uitkomste van vroue en kinders vas te stel. Die IHS verskaf inligting oor die vraag of ’n huishouding by FISP baat of nie, wat die belangrikste behandeling van belangstelling in hierdie tesis is. Daar is ook ander bykomende data wat ek in spesifieke hoofstukke gebruik het om die IHS aan te vul om my navorsings vrae te beantwoord. Ek het die IHS byvoorbeeld gekoppel aan satellietdata om die impak van FISP op oesopbrengste te verifieer. Ek het ook die data vanuit die “Service Provision Assessment” opname, wat inligting oor gesondheids fasiliteite in gemeenskappe bevat, gebruik om te verstaan hoe FISP kombineer met moedersorg van goeie gehalte om kinder voeding te versterk. Ek gebruik hierdie data om in vier bre¨e doelwitte te verstaan of vroue besluite uitoefen oor die inkomste uit die FISP en of die kinders se uitkomste verbeter as gevolg van die FISP. In die eerste doelwit van die tesis heroorweeg ek die impak van FISP op mielie produktiwiteit, om die konteks van die drie ander doelwitte wat in die tesis gevolg is, te verskaf. In teenstelling met vorige bewyse wat opgesomde opname gebruik wat geneig is tot Nie-Klassieke Metings Foute, neem ek ’n stap verder om strategiese vooroordeel in die rapportering van opbrengste uit te skakel deur satelliet-gegenereerde data te gebruik. Ek het gevind dat die FISP gekoppel is met toenemende mielie produktiwiteit net in die suide van Malawi, ‘n streek wat die hoogste armoede vlakke, die minste bewerkbare grond en ho¨e bevolkingsdigthei, maar nie in die noord of sentrale streke met teenoorgestelde sosio-ekonomiese status. Ek argumenteer dat die verskille plaasgevind het, want in die suide het die FISP die arm huishoudings geteiken wat dit die nodigste gehad het om insette te bekostig. In die tweede hoofstuk het ek geondersoek of langtermyn-kindervoeding, lengte-vir-ouderdom, verbeter het onder kinders onder die ouderdom van vyf jaar in huishoudings wat die FISP ontvang het. Op grond van vorige evaluerings wat bevind het dat die FISP slegs die kort- en mediumtermyn voeding verbeter het, het ek die analise uitgebrei om die heterogene impak van die FISP op ’n langtermyn-voedings uitkoms vir kinders, hoogte-vir-ouderdom, te verstaan. Die resultate het aan die lig gebring dat die FISP die lengte-vir-ouderdom vir kinders onder die ouderdom van vyf jaar verbeter, slegs toe hul moeders blootgestel is aan “Focused Antenatal Care” (FANC) wat advies gee oor goeie dieet vir swanger moeders en hul kinders. Dit is waarskynlik dat die FISP die beskikbaarheid van voedsel verhoog het en likiditeit verskaf het wat die aankoop van meer voedsame voedsel moontlik gemaak het. Ek toon aan dat hierdie voedings beleggings net langer as drie jaar geduur het toe die kind blootgestel was aan opvolg deelname aan ’n gemeenskaps voedings program vir kinders onder die ouderdom van vyf. In die derde hoofstuk van die tesis het ek ondersoek of vroue meer geld het as kontant wat uit FISP-gesubsidieerde gewasse verdien word. Ek voeg ’n kontekstuele faktor by wat die sleutel is tot die bepaling van hoe vroue hul besluitnemings bevoegdheid in Malawi uitoefen-skikking spraktyke na die huwelik binne verwantskaps tradisies. My resultate toon dat die FISP nie die besluitnemings vermo¨e van vroue verbeter oor die verdienste uit mielie verkope nie. In werklikheid verskuif die program die besluitnemings bevoegdheid oor die verdienste van vroue na hul mans in matrilokale gemeenskappe. In matrilokale omgewings woon egpare in die geboorte gemeenskap van vroue, en die vroue besit huishoudelike grond wat beheer word deur hul oom. Ek toon aan dat die wins in die matrilokale man se besluitneming weens FISP groter is as die verlies in die besluitneming van vroue, wat lei tot die vermoede dat die bykomende wins vir mans afkomstig is van die oom van die vrou. Omdat mans namens hul huishoudings FISP ontvang, dui die resultate daarop dat mans in matrilokale nedersettings die program gebruik om hul ongunstige voordeel dat hulle nie grond besit nie, teen te werk. In die vierde hoofstuk het ek die impak van FISP op die uitval van kinders op skool geondersoek. Ek vind dat FISP die matrilokale meisies se skooluitval verminder. Die gevolge word grootliks aangedryf deur verhoogde uigawes aan meisies se opvoeding as gevolg van die program. Matrilokale ouers het waarskynlik ’n deel van die opbrengs van FISP in meisies belˆe, want in die toekoms na die huwelik het matrilokale vroue saam met hul ouers gesorg en sodoende vir die ouers gesorg as hulle ouer word - ’n plig wat seuns in patrilokale gemeenskappe onderneem. Die belegging in meisies vir toekomstige voordele van ouers, is meer relevant vir matrilokale mans omdat hierdie mans nie grond besit nie. Selfs sonder om grond te beheer, kan die mans dus ten minste sorg van hul dogters ontvang. Ek vind dat FISP nie skooluitval in patrilokale gemeenskappe be¨ınvloed nie. Vanuit ’n Malawiese konteks van algehele uitval van ho¨erskole, is hierdie resultaat waarskynlik omdat seuns, wat erfgename is in patrilokale gemeenskappe, reeds ’n laer uitval het. Ek kom tot die gevolgtrekking dat FISP assosieer met verbeterde mielieproduktiwiteit wanneer dit op arm mense gerig word, en FISP jonger kinders bevoordeel met verhoogde voeding. Die FISP verminder die geslags ongelykheid slegs onder skoolgaande kinders in matrilokale gemeenskappe, terwyl dit geslags ongelykheid in besluitneming onder ouer vroue in dieselfde gemeenskappe verhoog. Alternatiewe beleide, wat bewys het dat dit die sleutel is tot die verbetering van vroueagentskap in ander lande, soos direkte kontant oordragte aan vroue, moet bevorder word, want dit is onwaarskynlik dat dit sal toeneem om die agentskap te verhoog deur die FISP direk op vroue te rig. Dit is omdat beheerders van grond en die opbrengs daarvan is mans-die vrou se man in patrilokale gemeenskappe, en ooms en eggenote in matrilokale gemeenskappe. Die algemene welsynsbeleid, soos FISP, moet die tradisionele konteks van vroue en kinders verreken, om onbedoelde negatiewe gevolge, soos erger geslags ongelykhede, te vermy. Daar moet selfs welsyns trategie¨e wat spesifiek is vir die verbetering van die welstand van vroue en kinders, aangepas word vir kontekstuele faktore soos verwantskaps tradisies, om ’n wyd-verspreide positiewe impak op vroue en die welstand van kinders te hˆe.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/124557
This item appears in the following collections: