Dissertation Example-The effects of Conditional Cash Transfer (CCT) on child nutrition following an adverse weather shock: the case of Indonesia

Liliana Itamar Carillo Barba

 

Published: Jan 2023

Department of International Development
London School of Economics and Political Science
Houghton Street                                                                             Tel: +44 (027957425/6252
London                                                                                             Fax: +44 (020) 7955-6844
WC2A 2AE UK                                                                                  Email: [email protected]

 

Website: http://www.lse.ac.uk/internationalDevelopment/home.aspx

 

Abstract

 

Adverse weather shocks negatively affect health and child nutrition and the poorest populations in particular. Conditional Cash Transfer (CCT) programmes have been suggested to mitigate weather shock effects; however, little is known regarding child malnutrition. This study examines whether Program Keluarga Harapan (PKH), Indonesia’s flagship CCT programme, mitigated negative impacts of extreme rainfalls caused by the ENSO event (La Niña) in 2010-2011. Specifically, it uses an instrumental variable to evaluate the impact of PKH on stunting (low heightfor-age). Subsequently, it is interacted with a dummy variable that defines the weather shock. The results show that PKH is more effective at reducing stunting in the absence of the weather shock. However, PKH mitigates the negative effects of La Niña when the parents of a child are unemployed or illiterate, emphasising the potential of CCT for the most vulnerable.

 

 

1. Introduction

Child malnutrition is a global issue that primarily affects developing countries. It is estimated that 49% of deaths among children under five years of age are associated with undernutrition (World Health Organization 2021), which reveals vast health disparities. These inequalities between and within countries are reinforced by adverse weather shocks, perpetuating poverty (Gloede, Menkhoff, and Waibel 2015). Social protection programmes such as Cash Transfers (CT) aim to increase the wellbeing of their beneficiaries and have been suggested to mitigate the effects of negative weather shocks.

Among the several sub-forms of undernutrition, stunting is considered the best indicator of children’s health and wellbeing as it reveals chronic undernutrition. The adverse effects of stunting (low height-for-age) are not limited to childhood and persist through adulthood (World Health Organization 2021; de Onis and Branca 2016). The negative consequences of stunting not only threaten communities in the short term but also harm countries’ human capital in the long run, boosting social inequalities.

While stunting affects one-third of children under five years in developing countries (UNICEF 2007), extensive research has elucidated the factors and mechanisms that foster stunting, among the poor in particular (Vaivada et al. 2020). Consequently, it has been well established that retardation in linear growth is primarily caused by poor health settings, frequent infections, inadequate nutrition, and deficient care (de Onis and Branca 2016). It has also been suggested that weather variability, such as extreme precipitation and flooding, negatively affects children’s nutrition, hence causing stunting (Lieber et al. 2022).

Climate change has triggered a shift in precipitation patterns. As a result, extreme hazards such as flooding are becoming more severe, and climate events such as El Niño Southern Oscillation (ENSO) are expected to become more frequent and intense (Hirabayashi et al. 2013; Dimitrova and Muttarak 2020; Cai et al. 2015). Those events can also disrupt crop production, interrupt food supplies, disturb economic systems and damage human health in different pathways (Watts et al. 2018). People living in extreme poverty do not usually have the means and tools to prevent, cope with and adapt to weather impacts; consequently, they are affected the most (Väänänen et al. 2019).

Governments globally have employed various actions to protect their most vulnerable populations from negative weather shocks, including expansion of financial inclusion, natural disaster relief funds, and weather shock insurance (Hallegatte et al. 2017). However, these strategies may not be good at targeting the weakest. Conversely, some social programmes that target the poor, such as CT programmes that hand cash to the poorest aiming to smooth consumption, have shown significant positive effects on nutrition, health, and education. CT are also being studied through the lens of helping vulnerable households and communities cope with weather shocks (Wood 2011; Väänänen et al. 2019).

The literature evaluating the effect of CT programmes in alleviating adverse weather shocks has shown positive results regarding household resilience (Premand and Stoeffler 2020), wellbeing (Carraro and Ferrone 2020), welfare and food security (Haile 2021; Asfaw et al. 2017), children’s school attendance (de Janvry et al. 2006), and no increase in children labour (Fitz and League 2021). There are inconclusive results from the two largest CT programmes regarding children’s health. For instance, Bolsa Familia in Brazil (League and Dylan 2022) has shown positive effects, and PROGRESA in Mexico shows no influence (Aguilar and Vicarelli 2022). Therefore, the impact of CT in mitigating the effect of extreme rainfalls or flooding on children’s nutrition outcomes remains unknown.

In the last decade, the national prevalence of stunting in Indonesia has persisted at approximately 37% (Beal et al. 2018), which creates a fascinating setting to study. Additionally, the Government of Indonesia presented its first Conditional Cash Transfer (CCT) programme, Program Keluarga Harapan (PKH) in 2007 (Alatas 2011), inspired by successful CCT programmes in Latin America such as Bolsa Familia and PROGRESA. . During the PKH’s pilot programme (2007-2013), Indonesia was hit twice by ENSO cold events (also known as La Niña), namely in 2007-2008 and 2010-2011. The second has been the strongest ENSO cold event in the past eight decades and affected regional precipitation levels (Boening et al. 2012).

This study aims to answer the question: What are the effects of the PKH programme on stunting in children that experienced an adverse weather shock? Additionally, to explore the conditions in Indonesia, this study examines the effects of the PKH programme on stunting and the impact of the La Niña event in 2010-2011 on stunting. Therefore, the CT programme and the weather shock are explored independently and in conjunction. The hypotheses are that PKH reduces the levels of stunting and that a weather shock will increase these levels.

The framework for malnutrition and background of the impact of weather shocks and CT programmes are presented in section 2. Section 3 describes the datasets used for this study and their limitations. Furthermore, section 4 describes the empirical strategy, followed by an examination of the main empirical results in section 5. Section 6 aims to explain those results and identify underlying mechanisms. Additionally, robustness checks for these results are described in section 7. Finally, section 8 concludes this study.

 

 

2. Background and framework

 

2.1 Childhood undernutrition

 

Childhood malnutrition, commonly referred to as malnutrition, is a health condition resulting from reduced dietary intake, which causes nutritional disequilibrium. It hence disrupts the body’s normal functioning, but it is potentially reversible (Márquez-González and García-Sámano 2012). Malnutrition is an umbrella term for different categories of undernutrition predominant in children under five. The acute forms are underweight (low weight-for-the-age) and wasting (low weightfor-the-height). The chronic form of undernutrition is stunting, which is an indicator of linear growth, and stands for low height-for-age (World Health Organization 2021).

Stunting can begin in the first 1000 days after conception. It can start in utero and continue at least until the child is two years old. In this critical window, children who experience growth faltering can to catch up with their peers; once the 1000-day threshold is crossed, recovery becomes a slow process (Victora et al. 2010). Low height-for-age is associated with susceptibility to infections, where frequent, successive infections lead to worsening nutritional status, creating a vicious cycle difficult to break (Olofin et al. 2013). Child stunting is frequently associated with male children, premature birth, short maternal height, and nonexclusive breastfeeding. However, other factors such as low maternal education, low socioeconomic status, unhygienic household conditions, and rural areas are also closely associated with low height-for-age (Beal et al. 2018).

The United Nations Children’s Fund (UNICEF) developed a conceptual framework to further understand the social determinants and key indicators of maternal and child nutrition, where diets and care services are “immediate” determinants. Moreover, there are “underlying” and “enabling” determinants that play a role in these outcomes (Figure 1). The long-term consequences and intergenerational effects are also included in this framework (United Nations Children’s Fund (UNICEF) 2021; Lenters, Wazny, and Bhutta 2016).

 

The consequences of stunting have been profusely studied and are not restricted to childhood. In the short term, stunting increases mortality and morbidity, and increases the risk of diseases and infection, particularly diarrhoea and pneumonia (Black et al. 2013; de Onis and Branca 2016). There are also cognitive delays; stunted children have poorer cognitive ability and are more likely to get lower grades and drop out of school (Soliman et al. 2021). Low growth-for-age during adolescence is associated with higher risk of obstetric complications (Özaltin, Hill, and Subramanian 2010). In the long term, there is an increased risk of arterial hypertension, obesity, reduced height, and less developed cognition, memory and locomotor skills (Dewey and Begum 2011).

Stunted children mainly belong to low socioeconomic status. As mentioned before, they are less likely to develop their cognitive skills thoroughly, and as adults are more likely to have lower incomes. Additionally, those children whose mother was stunted in her childhood are at higher risk of being stunted themselves. Stunting hence contributes to the intergenerational cycle of poverty (World Health Organization 2018), underlining the importance of maternal nutrition especially during pregnancy.

Despite the efforts to fight malnutrition in recent decades, prevalence of stunting remains high across low- and middle-income countries (LMIC) globally (World Health Organization 2018; Victora et al. 2010). It reflects overall children’s wellbeing and reveals social inequalities (de Onis and Branca 2016).

 

2.2 Role of Cash Transfers in children’s nutrition

Malnutrition is a multifactorial entity; various underlying and enabling determinants are involved in the development of stunting, and several characteristics typical of poverty play an essential role. Therefore, programmes implemented in isolation might not reduce the stunting prevalence significantly. Tackling stunting programmes requires a holistic approach.

Some governmental strategies to reduce poverty and social vulnerability are social protection programmes that promote access to social and care services, social transfers and policies that guarantee equity (de Groot et al. 2015). These programmes help reduce chronic poverty, increase resilience among vulnerable households, support beneficiaries to meet their basic needs, and provide access to healthcare, nutritional support, and education. These programmes, specifically CT, have been accepted as an important strategy to promote child nutritionby directly and indirectly addressing various malnutrition-related factors (Ruel and Alderman 2013).

CT programmes address several immediate and underlying determinants of undernutrition. They provide cash to poor and vulnerable households upon the beneficiaries meeting certain conditions. Cash recipients are commonly required to use healthcare and prevention services and maintain school-aged children in school, aiming to enhance human capital formation and break the intergenerational transmission of poverty (Leroy, Ruel, and Verhofstadt 2009).

After the success of CCT programmes in Mexico and Brazil, CCT soon became popular among governments and stakeholders and were replicated in other LMICs. However, the transfer value and requirements for the beneficiaries vary according to the capacity and budget of each government. For instance, cash transfer programmes in Latin America (LA) are usually tied to behaviours described previously. In contrast, programmes are typically unconditional in SubSaharan Africa (SSA) (de Groot et al. 2017).

Extensive research has evaluated the results of CT on health, nutrition, education, wellbeing, spill overs, women empowerment, et cetera. Literature has increasingly found positive results regarding nutritional outcomes. Therefore it is worth understanding the mechanisms by which CT programmes reduce stunting. De Groot et al. (2015) established a framework to explain these. They propose that CT programmes affect underlying determinants via three elements: increasing food security, health (increasing health-seeking behaviours and provision of health care services) and care (health education and women empowerment). On the intermediate determinants level, CT programmes help increase child’s dietary intake and improve child health status (Figure 1).

For instance, it has been documented that CT improve household consumption in SSA and Latin America, showing increased intake of fruits, vegetables and protein-rich foods (de Groot et al. 2017). These programmes have also positively affected local markets by increasing demand of basic products (Mohammadi-Nasrabadi 2016). Moreover, children and pregnant women from households receiving CT benefits are more likely to attend preventative healthcare visits and general increase usage of the health care services (Lagarde, Haines, and Palmer 2009). In terms of care, the mental health of the caregiver is a relevant factor for children wellbeing, and it has been found that CT have a small impact on subjective wellbeing and mental health (McGuire, Kaiser, and Bach-Mortensen 2022).

As opposed to the underlying determinants, the intermediate determinants are factors concerning the individual’s condition. There has been limited research to explore whether CT improve child dietary intake, and given the diversity of programmes, results are mixed (de Groot et al. 2017). Furthermore, it has been found that children beneficiaries of CT are less likely to be ill and indicate better health overall (de Groot et al. 2017). Finally, CT have been estimated to decrease diarrhoea incidence (Manley, Alderman, and Gentilini 2022).

Overall, CT reduce stunting prevalence by 1.4%, and the impacts on height-for-age z-score (HAZ) scores are small but statistically significant (Manley, Alderman, and Gentilini 2022). However, the child’s age seems to play an important role; the effect appears larger in children under two years old. Each CT programme meets different criteria, length, conditionalities, and transfer size. Those characteristics should be considered when comparing CT programmes to each other.

 

2.3 Adverse weather shocks

The impact of weather shocks on human health has been studied for decades. For example, exposure to rainfall and temperature variability early in life has adverse effects on cognitive development and education (Adhvaryu et al. 2018), worsens mental health (Adhvaryu et al. 2015) and increases risk of unemployment (Adhvaryu et al. 2018). Remarkably, exposure to extreme rainfall and flooding significantly increases the risk of stunting (Lieber et al. 2022; Phalkey et al. 2015). Individuals who experienced a weather shock in utero or during childhood preserve those negative effects through adulthood (Cornwell and Inder 2015).

Sudden weather-related disasters might affect children’s health in three ways: 1) physical health, including lack of food security, increased of infectious diseases, contaminated drinking water and disrupted access to health care; 2) mental health; and 3) abandoning education and being forced into labour (Kousky 2016). Most of these factors can be integrated into UNICEF’s framework for child nutrition (Figure 1).

Variation in rainfall and flooding affects crop harvest and decreases the productivity of the land, hence causing shortages of food (Phalkey et al. 2015). Consequently, regional food systems may increase food prices, leading to food insecurity for the poorest households (Brown and Kshirsagar 2015). However, weather-induced disruptions to children’s nutrition can also occur even if there is no disruption of agriculture (Thiede and Gray 2020). For instance, disasters can interrupt access to medical care because of damage to clinics or road infrastructure (Kousky 2016). Also, it has been found that flooding leads to water contamination and therefore water insecurity (Rosinger 2018). Similarly, the increase of rainfall boosts the prevalence of vector-borne diseases such as malaria and dengue (Campbell et al. 2015). Additionally, shocks may alter domestic roles and labour supply. In rural contexts with gendered division of labour, the demand for women’s labour is increased and reduces the time available for childcare (Thiede and Gray 2020).

It is worth noting that the groups most vulnerable to natural hazards are those that have high exposure to them. In general, they tend to be the poorest, women, children, and marginal communities (Tirado et al. 2013).

Finally, climate change has caused an alteration in precipitation patterns and increased the frequency and magnitude of floods (Hirabayashi et al. 2013). Climate phenomena such as the El Niño Southern Oscillation (ENSO) are notable fluctuations in the climate worldwide causing tropical rainfall variability. ENSO has two conditions: cold (La Niña) and warm (El Niño). While this event occurs intermittently (every 2-7 years), it has had devastating consequences for agriculture and infrastructure, and displaced many people in vulnerable areas. Predictions are that climate change will only increase the frequency and intensity of this phenomenon (Cai et al. 2015; Rodysill et al. 2019; Tang et al. 2018).

 

2.4 The function of Cash Transfers in mitigating weather shocks

Weather-related disasters have direct economic effects on households that rely on farm activities for income; this increases their vulnerability to droughts and floods. The most evident consequences are the destruction of property, cattle, productive assets and infrastructure (Wisana 2015). Poor households do not have the means to cope with shocks. Additionally, they receive less support from financial instruments, private remittances and social protection programmes (Hallegatte et al. 2017). They have developed mechanisms to smooth their consumption, such as increasing family labour supply, reducing investments in education and health, selling assets, and activity diversification. However, these strategies do not promote sustainable development and create a vicious cycle of poverty (Schäfer et al. 2016).

There are numerous instruments used to protect households at risk of weather shocks. One of them is enhancing financial inclusion, aiming to diversify vulnerable households’ assets and economic activities. For instance, credit promotes a fast reconstruction of assets, private insurance against disasters offers protection to families at risk. However, these tools have substantial limitations among the poorest. CT have been contemplated as a way to compensate for those limitations and improve resilience among the poorest (Hallegatte et al. 2017).

Wood (2011) postulates a framework by which CT can provide relief in weather-related shocks, and this is adapted into the child nutrition framework (Figure 1). CT can support vulnerable households by 1) helping them meet existing needs and reducing the impacts of weather shocks since CT cover basic needs, the provision of food security and access to health care is ensured; 2) CT provide the financial means to endure shocks; 3) CT reduce the pressure for households to engage in negative strategies, such as cutting budget for healthcare. Furthermore, Carraro and Ferrone (2020) propose that CT can influence children’s nutritional status directly via the income effect or indirectly via the household’s coping capacity (reducing food security, increasing child labour or promoting migration of the working age members).

The literature evaluating the effect of CT programmes in alleviating shocks has shown positive results. Children’s school attendance was not disrupted by unemployment or illness of family members in the presence of CT (de Janvry et al. 2006). Multidimensional deprivation is not affected in the case of a self-perceived shock (Carraro and Ferrone 2020). In the case of weather shocks, CT have positively impacted household resilience (Premand and Stoeffler 2020), welfare, food security (Haile 2021) and consumption smoothing (Vicarelli 2011) in the case of droughts. Other studies assessing both edges of extreme rainfall variation (wet and dry) have found that CT increase food security (Asfaw et al. 2017), improve school grades, and prevent children from dropping out of school and moving into labour (Fitz and League 2021).

However, few studies evaluate the impact of extreme rainfall and floodings on children’s health, and their results are inconclusive. A study examining the effects of Bolsa Familia in Brazil showed positive effects on reducing stunting, obesity, and other adverse health outcomes. It is also found that the CCT is most effective on children when the shock happened in utero (League and Dylan 2022). Conversely, a study evaluating the effect of extreme rainfall caused by the phenomenon “El Niño” on beneficiaries of PROGRESA in Mexico shows that children’s cognitive development is not protected by the CCT program (Aguilar and Vicarelli 2022). Thus, the effect of CCT in mitigating the impact of negative weather shocks such as extreme rainfall or flooding on child nutrition remains unclear.

 

2.5 Indonesia – case study

One region in the world that has high prevalence of stunting is East Asia and the Pacific, where 21 million children under five years old are stunted (Drummond, Watson, and Blankenship 2021). Among those countries, Indonesia shows a prevalence of stunting much higher than other countries (Soekatri, Sandjaja, and Syauqy 2020). Despite the efforts to fight malnutrition, high disparities between provinces persist and the national prevalence of stunting continued at approximately 37% in the recent decades (Beal et al. 2018).

The high prevalence of stunting in Indonesia is also reflected in the country’s poverty incidence and vice versa. In 2007, 16.65% of inhabitants lived in relative poverty, and 39 million citizens lived in absolute poverty, primarily in rural areas (Aji 2015). In that same year, the Government of Indonesia presented its first CCT programme, Program Keluarga Harapan (PKH), to improve the wellbeing of the poorest. The PKH programme was assessed using a randomised control trial. Three hundred sixty subdistricts evenly divided into control and treatment groups were surveyed. Follow-up surveys took place in 2009 and 2013 (Alatas 2011).

During the PKH’s pilot programme (2007-2013), Indonesia was hit twice by the ENSO phenomenon “La Niña”, in 2007-2008 and 2010-2011. The second has been the strongest ENSO cold event in the past eight decades and affected regional precipitation levels (Boening et al. 2012). In Indonesia, the drier conditions of the ENSO phenomena occur during El Niño and wet conditions during the La Niña (The World Bank 2021). Given its location on the equator, precipitation is the element of weather that has the most variation in Indonesia, and the temperature barely fluctuates (Levine and Yang 2014). Therefore, this weather shock is associated with rainfall only.

The consistent high prevalence of stunting, the PKH pilot implementation, and the La Niña event create a natural experiment in Indonesia. This serves to evaluate the effect of extreme rainfall caused by La Niña on CCT beneficiaries, particularly stunted children.

 

3. Data and descriptive statistics

The present study uses two large datasets to evaluate heterogeneous impacts of CCT programmes on children’s health in populations affected by negative weather shocks. The datasets are Indonesia’s CCT’s pilot dataset and the monthly precipitation per district data.

 

3.1 PKH programme

In 2007, the Government of Indonesia launched its first CCT programme, PKH. This social programme aimed to reduce poverty by improving the welfare of extremely poor households. The nature of its design promoted low-income families to escape intergenerational poverty (Alatas 2011). In addition, the initiative stimulated the development of human capital (especially children and pregnant women) by encouraging beneficiaries to attend school and visit health clinics regularly (Dulkiah, Avid, and Irwandi 2018).

The PKH’s pilot program was implemented in seven provinces , and this selection ensured diversity of the regions for national representativity. The targeted regencies (kapubaten) were those with a high incidence of poverty and malnutrition and a low transition rate of education from primary to secondary school. Districts (kecamatan) that were “supply-side ready” were randomly selected for participation in the pilot program (Alatas 2011); 438 subdistricts were assigned to treatment and 298 to control. Consequently, targeted households were those listed in other social programme catalogues and classified as extremely poor using a proxy-means test. Lastly, households that met the eligibility criteria such as having children aged 0-15 years, and/or children less than 18 years old but that had not yet completed nine years of basic education, and/or pregnant/lactating women, were included in the pilot programme (Sparrow 2008).

The female head of the eligible household received quarterly cash transfers based on the number of children, their ages, and if there was a pregnant or lactating woman in the household. The maximum transfer per household was 2,200,000 rupiah (approximately 150 USD today) per year. It was aimed to provide the equivalent of 15-20% of the estimated yearly consumption of poor households (Alatas 2011).

 

The administrative subdivisions of Indonesia are provinces composed of regencies (kabupaten).
Regencies are divided into districts (kecamatan), these are made of villages (desa or keluharan).

 

The raw survey data of the PKH’s pilot is publicly available, provided by Cahyadi et al. (2020). Data regarding children’s health, household and village characteristics of control and treatment groups are used for this study. This data set collects the baseline survey before the CCT was implemented (2007), the evaluation survey by the World Bank two years after (2009), and a third survey wave by Cahyadi et al. (2020) following the same structure as the former waves. Household attrition was low; 95.1% of the household surveyed at baseline were still found in the last survey (Cahyadi et al. 2020) (See section 7.4).

At baseline, the final sample is balanced across a wide range of key variables between the control and treatment groups. Column 4 of table 1 shows the average differences between these groups. The standard error is clustered by sub-district according to the level of randomisation as suggested in the literature (Colin Cameron and Miller 2015). None of these variables show statistically significant differences at the 10 per cent level, suggesting no difference between the groups.

 

The main outcome of interest is stunting, which is a widely used indicator for impaired growth and children’s development (World Health Organization 2015). Height-for-age z-score (HAZ) is calculated according to the age and gender of the child using STATA macro “igrowup” (World Health Organization 2022). Z-scores in malnutrition are used to illustrate the individual status of the child in relation to the population used as the reference standard. Moderate stunting takes zscore values between -2 and -3, and severe stunting less than -3 (Mehta et al. 2013). Even though stunting is the primary outcome variable, HAZ scores are also explored to visualise the spectrum for low height-for-age, as described by Ledlie et al. (2018).

This extensive panel data set is therefore fundamental for this study. Table 2, panel A shows summary statistics for the most relevant variables of the study sample: those who were CCT beneficiaries by the end of the PKH’s pilot programme.

 

 

3.2 Precipitation data

Extreme rainfall impact on PKH beneficiaries is evaluated in this study using a dummy variable that is constructed as follows.

Firstly, monthly precipitation data is provided by the Climate Research Unit Time Series (CRUTS) of the University of East Anglia (Harris et al. 2020), which provides total monthly precipitation data on a 0.5o latitude by 0.5o longitude resolution statistically interpolated. This gridded data has the advantage that it does not contain any missing observations at certain locations (Auffhammer et al. 2013). Data from 1980 to 2013 is spatially matched to Indonesian districts and extracted using IPUMS-Terra (Ruggles et al. 2018).

This study uses the “standardised anomaly” or Z-score index (ZSI), a versatile tool that analyses wet and dry periods and shows the extent to which precipitation varies from its average state. It compares a given observation to the same location’s long-run average precipitation and is normalised by the standard deviation (Y. Li et al. 2019). It is calculated using the following equation:

 

Where 𝑋𝑖 is the observed precipitation in district d, in year t; 𝑋̅ is the long-term average; and 𝜎 is the standard deviation of the long-term frame (Noor et al. 2020). For the wet spectrum, a ZSI > 1.44 is considered wet, while ZSI > 1.96 is extremely wet (X. Li, Zhang, and Ye 2013). This index has been used inconsistently in the literature. Some studies use a standard deviation above one to identify positive rainfall shocks(League and Dylan 2022). In contrast, other studies use this same threshold to define flood shocks (Carrillo 2020; Adhvaryu et al. 2018). Lastly, a study evaluating extreme rainfall on corn production defines extreme rainfall conditions as a precipitation anomaly of 2.5 (Y. Li et al. 2019). As a result, this study uses an arbitrary ZSI of >2 as a threshold to determine an adverse rainfall shock, to avoid including positive effects of precipitation.

Note that when calculating the annual rainfall average, this study uses wet-dry season years instead of calendar years, which is closely related to agricultural cycles. Since the primary production of rice and corn is aligned with the seasonality of the rain (International Production Assessment Divison n.d.), the significant link between rainfall and crop production has been well documented (Naylor et al. 2007). For that reason, in this study, the “year” starts with the first month of the “dry” season of a given province and finishes the last month of the wet season. The definition of season year per province is provided by Maccini and Yang (2009).

The last cold ENSO event categorised as ‘strong’ happened in 1999-2000. It is assumed that if there was rainfall variability caused by the La Niña in 2010-2011, it will be reflected in the ZSI. Therefore if 2010 or 2011 season years had a ZSI >2, a value of 1 is assigned to the “weather shock” variable. This new variable is set at the district level, and it is matched to the individuals according to the district of residence, available in the PKH dataset.

Table 3 shows the balance test for the individuals who experienced the weather shock and those who did not. Column 1 shows the summary statistics of the most relevant variables for the population who suffered the weather shock, and column 2 shows those who did not. None of the variables are statistically significant. Consequently, it can be assumed that a weather shock is a random event. Panel B of table 2 shows that weather shocks affected the population in the PKH pilot, regardless of the assignment of treatment or control.

 

3.3 Data pitfalls and assumptions

The WHO measures malnutrition z-scores for children up to 5 years old (World Health Organization 2006). Thus, low height-for-age z-scores obtained in the last survey represent children born between 2009 and 2013 and those children screened in 2007 are not included. However, the pathways in which weather shocks affect children’s health and nutrition do not necessarily need to occur in the post-birth stage. Therefore, this study assumes that the weather shock that happened in 2010-2011 also affects those born after. Figure 3 visually explains the order of events.

Moreover, enrolment into the PKH programme did not strictly follow the initial random assignment. First, there were further household screenings to determine PKH eligibility after randomisation. Additionally, by 2013 the programme was active in 99% of the treatment and 39% of control group (Cahyadi et al. 2020). Therefore, enrolment into the PKH programme may be correlated with other confounders at the household-level. Thus, this variable may suffer from endogeneity, and additional treatment is necessary.

Additionally, the verification systems in charge of validating compliance and recalculating benefits (including new-borns, excluding those who turned 18, etc.) were not operating until 2010; therefore, compliance was not enforced. Nevertheless, the World Bank report states facilitators ensured that the beneficiaries acknowledged the conditions and were aware of the risk of losing their payments (Alatas 2011). However, Cahyadi et al. (2020) state that conditions were not always enforced even after verification systems were functional. The inconsistent verification of compliance could lead to different behaviours among beneficiaries. Yet, “labelled cash transfers” have been found to yield similar outcomes to those programmes that actively enforce conditionalities (Heinrich and Knowles 2020; Benhassine et al. 2015). This study assumes that the roll-out of verification systems and facilitators’ skills to share the purpose of the cash was uniform in all villages.

The PKH data set does not provide names and locations of villages and districts to safeguard the identity of the respondents; therefore, the rainfall must be matched at the regency level. Additionally, rainfall data might suffer from aggregation bias which suggests that such temporal and spatial aggregation might not reflect the weather conditions appropriately in the entire district (Auffhammer et al. 2013). However, other studies in Indonesia have faced this issue yet found a significant effect of rainfall on children’s health (Cornwell and Inder 2015).

 

4. Empirical strategy

First, this study evaluates the effect of the PKH programme on children’s stunting. Then, it explores the impact of extreme rainfall on stunting in Indonesia. Finally, it examines the capability of PKH to mitigate weather shock effects on stunting.

 

4.1 Effect of the PKH

As explained above, the variable that accounts for receiving the PKH benefits suffers from endogeneity, given that compliance with the original randomisation assignment was not completely satisfied. Thus, an instrumental variable (IV) addresses this endogeneity. Initial random allocation is used to instrument the enrolment into the PKH programme as specified in the evaluation of the PKH pilot programme (Cahyadi et al. 2020) and suggested for randomised experiments elsewhere (Angrist and Krueger 2001).

Therefore, the correlation between stunting and receiving PKH’s benefits is estimated using a 2- Stage Least Squares (2SLS) approach. To assess if a child i, in household h, in a village v, in subdistrict s, in district d, is a PKH recipient (𝑃𝐾𝐻𝑖ℎ𝑣𝑠𝑑), the regression for the first stage is as follows:

Where 𝑅𝑎𝑛𝑑𝑜𝑚𝑖𝑠𝑎𝑡𝑖𝑜𝑛𝑠 stands for the original randomisation assignment and indicates whether the household was assigned into the treatment group. To increase the precision in the results, individual controls (𝑋𝑖 ), household controls (𝑋ℎ) and village controls (𝑋𝑣 ) are included. Individual controls account for the age and gender of the child. Household controls and village controls comprise variables that may influence children’s health (social determinants of health); these values are obtained from the first wave, before the pilot programme was implemented as suggested in the literature (Cahyadi et al. 2020). Such controls are the natural logarithm for household size and expenditure, whether the caretaker is literate and is employed in agriculture, access to clean drinking water, and whether the household has access to electricity and improved latrine. At the village level, the average of stunted children, and whether the village is considered rural or urban are variables controlled for. 𝜇𝑑 are regency (kabupaten) fixed effects that control for potential additional programmes, policies, and other state factors. Standard errors are clustered at the district level, which is the level of randomisation of the programme (Colin Cameron and Miller 2015).

The estimation of 𝑃𝐾𝐻𝑖ℎ𝑣𝑠𝑑 is used in the second stage regression as follows:

Where 𝑌𝑖ℎ𝑣𝑠𝑑 is the dummy variable “stunting” that takes a value of 1 if the HAZ score is between -2 and -3 standard deviations and 0 otherwise, of a child i, in household h, living in a village v of subdistrict s, of district d, as. The primary outcome to be analysed is stunting; however, extreme stunting and HAZ scores are also explored to account for changes in each category (Kandpal et al. 2016). The same controls as in the first stage are used for consistency and precision.

 

4.1.1 Instrument validity

Two assumptions must be met for an instrument to be valid, namely 1) it must be relevant and cause variation in the treatment variable, and 2) it should not have a direct effect on the dependent variable (only via the treatment variable). This allows the researchers to estimate the level of exogenous variation (Newhouse and McClellan 1998).

The first assumption proves valid in table 4, showing the first stage results. It is found that households initially assigned to the treatment group are 40.5% more likely to have received the PKH benefits after six years. The high value of the F-test provides the confidence to reject the null hypothesis that the instrument is weak.

The second assumption implies that the instrument must be exogenous, and that the randomisation assignment should not be correlated with the prevalence of stunting but only via CCT receipt. Here, there are two main potential concerns: across and within districts spillovers.

During the PKH’s pilot sample selection, subdistricts that statistically had sufficient health and education services were included in the sample (Sparrow 2008), therefore those services are contained within each subdistrict, and spillovers are not likely to occur in this context. Randomisation was performed at this level for this reason (Cahyadi et al. 2020). Spillovers across subdistricts are also unlikely in this setting given the specific targeting of the PKH programme. Hence a small portion of households in the standard village assigned to control were treated (6.5% of households). General equilibrium effects such as overcrowding health clinics and schools are likely to be small in this scenario (Cahyadi et al. 2020).

These arguments support the two assumptions to corroborate that initial randomisation assignment into the PKH’s pilot program is a valid instrumental variable for PKH treatment.

 

4.2 Effect of adverse weather shocks

The evaluation of the impact of weather shocks on the health of children under five years using the following regression for child i, in household h, in a village v, in subdistrict s, in district d:

Where 𝑌𝑖ℎ𝑣𝑠𝑑 is the health outcome as defined before. 𝑆ℎ𝑜𝑐𝑘𝑑 is a dummy variable that takes the value of 1 if the district suffered from extreme precipitation in 2010-2011 and 0 otherwise. Extreme precipitation is considered if the ZSI >2. The controls and fixed effects are used as in the previous models in this study.

 

4.3 Full model

To assess the interaction between receiving the benefits of PKH and the negative weather shock on children’s nutrition, the full model is used:

Where, 𝑌𝑖ℎ𝑣𝑠𝑑 represents if a child i living in in household h, in a village v, in subdistrict s, in district d is stunted. 𝑆ℎ𝑜𝑐𝑘𝑑 represents the dummy variable explained above. 𝑃𝐾𝐻 ̂𝑖ℎ𝑣𝑠𝑑 is the instrumented PKH receipt. 𝑆ℎ𝑜𝑐𝑘𝑑̂𝑃𝐾𝐻𝑖ℎ𝑣𝑠𝑑 represents the interaction term between rainfall shocks and receiving PKH benefits. The standard errors are clustered at the regency level, the highest level of randomisation defined by rainfall.

𝛽1 coefficient measures the effect of the rainfall shock on stunting on those who were not CCT beneficiaries, which is expected to be positive (i.e. increase stunting). Whereas 𝛽2 coefficient represents the effect of receiving PKH on the outcome variable and that population did not experience the weather shock, which is estimated to be negative..

Lastly, the coefficient of interest, 𝛽3 estimates the additional effect of receiving cash transfer in the presence of a negative rainfall shock. 𝛽3 > 0 indicates that PKH is more effective on those who had not experienced the La Niña 2010-2011, and 𝛽3 <0 indicates that PKH is more effective on those who were affected by the weather shock. Additionally, if 𝛽1 + 𝛽3 > 0 there is partial recovery of stunting and if 𝛽1 + 𝛽3 > 0 there is full recovery.

 

4.3.1 Proving similar conditions

This study assumes that the weather shock is a source of exogenous variation, given that this event was the strongest in Indonesia in decades. The balance test of the weather in table 3 and Panel A of table 2 demonstrates that the group that experienced the weather shock has similar characteristics to the group that did not, affecting equally the treatment and the control group of the PKH’s pilot programme.

To increase the robustness of this assumption, a t-test (Table 5) was performed on the HAZ score in the sample at baseline. The p-value lower than 0.05 demonstrates that there were no significant differences in baseline between the group that was later affected by a weather shock and the one that did not. Furthermore, the distribution of the HAZ score is normal and similar in those who experienced the weather shock and those who did not (Table A1).

 

Additionally, both groups followed almost parallel trends regarding HAZ levels through 2007 and 2009 (Figure 4). However, there are not enough data points to create a robust difference in difference identification strategy.

 

Thus, there were similar child nutrition conditions at the baseline and during the second survey in both groups. These considerations provide strong arguments to suggest that the weather shock is a source of exogenous variation, and it is highly correlated to nutritional outcomes.

 

5. Results and discussion

This section describes the empirical results of the strategy previously described. The first section presents the summary statistics of stunting. The second section explores the isolated effect of the PKH programme on the outcome variable, followed by the isolated effects of negative rainfall shocks on stunting. Finally, section four evaluates adverse weather shocks’ interaction with CCT recipients.

 

5.1 Low height-for-age in Indonesia

Several studies have identified the risk factors associated with stunting in Indonesia (Beal et al. 2018; Torlesse et al. 2016). These factors are briefly explored to understand the heterogeneity of the results of this study.

The summary statistics for stunting at the end of the pilot programme are shown in table 6. It is found that the oldest children in the sample are the most stunted. Additionally, children living in households where the caretaker is literate and employed in the agricultural sector are less stunted than children from an opposite situation. Infants that do not have access to an improved latrine, clean drinking water or electricity services are more stunted than children with better household conditions. These findings are consistent with other studies in and outside Indonesia (Beal et al. 2018; Torlesse et al. 2016; Utami, Setiawan, and Fitriyani 2019) and other LMICs (Amadu et al. 2021; Danaei et al. 2016).

The child’s gender is relevant as girls are less stunted than boys. This result is expected given that boys are more vulnerable to infectious diseases such as respiratory infections, diarrhoea, and malaria and are more likely to be preterm at birth, increasing male morbidity, mortality, and weight loss. Also, boys have a higher calorie that requires more resources to meet (Wamani et al. 2007). However, how those biological mechanisms interact with social conditions and affect stunting has not been sufficiently studied (Thurstans et al. 2020).

 

5.2 Effect of the PKH programme

The empirical strategy in equation 3 aims to explore the effect of the PKH pilot programme on child nutrition. Table 7 shows the results of the 2SLS.

It was found that stunting and severe stunting are reduced significantly by the CCT. For example, receiving PKH reduces stunting by 7 percentage points, while severe stunting does so by 9 pp, with a confidence level of 90 and 95%, respectively (columns 1 and 2). Similar results were found in the Philippines’ flagship CCT programme, where there was a higher significant reduction in severe stunting than stunting (Kandpal et al. 2016).

Moreover, HAZ increased by 0.37 units for children who received cash transfers. When disaggregating HAZ by sex (columns 4 and 5), it is found that girls benefit more than boys. This result is statistically significant and is consistent with the literature (Manley, Gitter, and Slavchevska 2012).

The first 1,000 days of life are a potentially critical window for chronic undernutrition; this time frame corresponds from the period in utero until the child is 24 months old. Table 8 disaggregates the effects of stunting by age in months. Columns 1 and 2 show the first and second years of life, and it is found that CCT reduce stunting by 7 and 5 percentage points, respectively. Nevertheless, these results are not statistically significant.

Column 3 shows that children just above the critical window (25-36 months old) do not benefit from the PKH to reduce stunting. This could be associated with the remark that interventions outside this critical window may not have meaningful effects (Ruel 2010). Nevertheless, Leroy and Frongillo (2019) argue that delays in linear growth continue after the first 1,000 days and that some cognitive skills that are dependent on nutritional status keep developing until adolescence. Therefore, special interventions should be addressed for this age group to ensure healthy growth. Children that are four and five years old are still helped by the CCT, albeit to a lesser extent than their younger peers.

5.3 Effect of the adverse weather shock

The effect of the cold ENSO event in 2010-2011 on children’s nutritional outcomes is evaluated using equation 4. The results are displayed in the following table.

The weather shock decreases HAZ by 0.03 units, which is reflected in an increase of stunting of 14 percentage points. Though not significant, these results align with Tanzania’s literature that evaluated extreme rainfall effects on stunting (Nsabimana and Mensah 2020). Counterintuitively, extreme rainfall seems to positively affect severe stunting, reducing it by 0.04 percentage points. However, these results must be treated with care as they are not statistically significant.

Additionally, the results for stunting are disaggregated by sex and location of the villages (columns 4-7). Female children who experienced a weather shock increased stunting by 20 percentage points (statistically significant at 90%), while stunting in male children rose by 10 percentage points (not significant). These results suggest that girls are more affected by extreme rainfall shocks, which is consistent with a study in India that analyses the effect of extreme precipitation, where stunting increases in girls exposed to the weather shock (Dimitrova and Muttarak 2020).

Furthermore, experiencing the weather shock increased stunting by 18 percentage points in rural villages (statistically significant). This is consistent with literature studying seasonal droughts in Ethiopia (Dimitrova 2021) and cumulative exposure to dry weather shocks in Tanzania (Nsabimana and Mensah 2020). However, in urban settings, weather shocks seem to decrease stunting by 26 percentage points, proposing a positive effect on stunting reduction, though this result is not statistically significant. These results are compatible with earlier findings by Nsabimana and Mensah (2020), who found that shocks have little and even positive impact on stunting when evaluating urban areas.

To further investigate the effect of the La Niña in 2010-2011. The results are disaggregated by age in months (Table 10). For babies up to 12 months, the weather shock is associated with a decrease in stunting, suggesting a positive effect (column 1). On the other hand, extreme rainfall increases stunting in children 13 months old and older (columns 2-5).

Children aged 3-4 years old at the time of the survey (which took place between September and December 2013) were in their first or second year of life when the La Niña occurred. For them, stunting increased significantly by 33 percentage points (column 4). Indeed, literature finds that weather shocks occurring during the first year of life have the strongest effects on children’s outcomes, particularly height (Maccini and Yang 2009; Rabassa, Skoufias, and Jacoby 2014; Nsabimana and Mensah 2020) or within the first 24 months (Dimitrova 2021).

Children between 4 and 5 years old (column 5) could have experienced the effect of the La Niña 2008-2009 in utero or during their first year of life. Despite knowing that event was less strong, a new dummy variable is created where 1 is assigned if the district had ZSI>2 during 2008 and 2009, and otherwise 0. However, this phenomenon is not reflected by this dummy variable as shown in Table A2. Therefore, it is assumed that the effects of that event are not observed in this sample.

 

5.4 Effect of the PKH programme and weather shocks

Finally, the full model accounts for the interaction between weather shocks and the impact of the CCT on stunting. Equation 5 is performed to explore this relationship. Results from the second stage are shown in table 11.

The effects on stunting are shown in column 1. Those who experienced the weather shock and did not receive CCT had increased stunting by 1 percentage point, yet this change is minor and insignificant. In contrast, receiving CCT benefits and not enduring a weather shock reduced stunting by 15 percentage points. These two results align with the hypotheses of this study. The interaction of these two treatment factors increases stunting by 11 percentage points, suggesting that PKH is more effective against stunting when the beneficiaries experience no weather shock and mitigates weather shocks poorly. However, none of these results is statistically significant and should be treated with care.

The difference between girls and boys has been discussed prior and appears to play an essential role in this interaction. Columns 2 and 3 show these results disaggregated by sex (statistically insignificant). Consistent with earlier results of this study, girls are the most affected by weather shocks in households that are not beneficiaries of CCT, whereas boys seem to benefit from this situation and benefit more from the PKH in case of an extreme weather shock than girls. Indeed, it has been recognised that in crisis settings coping mechanisms occur in families prioritising boys, and women and girls may suffer from a lower food intake than their male counterparts (InterAgency Standing Committee (IASC) 2017). However, this data set does not contain individual food intake to explore this mechanism.

The previous section indicated that rural areas are negatively affected by the La Niña. To isolate this effect further, table 12 stratifies the results by the age of children living in rural villages.

Consistent with previous results of this study, weather shocks play a positive role for babies <12 months old, reducing stunting by 15 percentage points (significant at 90%). Similarly, children aged 2-3 years are not helped by the PKH. The interaction terms are ambiguous and unclear; hence the interaction term has no heterogeneous effect on stunting by age. A potential explanation is that those children were not born, nor in utero by the time of the shock.

Overall, the PKH programme appears to be more effective (reduces stunting more) for those children who did not experience the weather shock compared to those that experienced the La Niña (Figure 5), independent of the sex of the child.

 

 

6. Potential mechanisms

Thus far, this study has found that that PKH did not help mitigate the negative effects on nutrition caused by the La Niña in 2010-2011.

Moreover, there are characteristics of the population that may enhance the efficiency of the PKH in the lack of a weather shock (Table A3). For instance, rural households with access to clean water and improved latrines show more significant improvements in stunting when they do not experience a weather shock. However, in case of weather shock, households with access to clean drinking water benefited from this feature, acting as a protective mechanism against extreme rainfall. These characteristics highlight the importance of hygiene and sanitation in reducing stunting in children under five years old, as described elsewhere (Ademas et al. 2021; Torlesse et al. 2016).

Conversely, PKH was effective at mitigating the effects of La Niña in rural households where the head was unemployed. CCT allowed children to catch up fully (B1+B3<0). Emphasising the capacity of CT as a form of insurance against risk in those without a stable income. A similar situation occurs when in a rural setting the caretaker was illiterate. Nevertheless, these results are not statistically significant (Table A4), and more research is required to comprehend these mechanisms fully.

As proposed in the UNICEF framework, household environment and sanitation are essential components for children’s nutrition and health. Extreme rainfall can put those elements at risk (Bhavnani et al. 2014). To explore the characteristics that play a role in this setting, additional analysis is done. It is found that children living in household with no electricity access benefit more from PKH, but have higher stunting after a weather shock, which could be related to the overall decreased wealth of the village (Fujii, Shonchoy, and Xu 2016). Additionally, children exposed to floors made from soil have higher chances of being stunted in case of extreme rainfall. Soiltransmitted helminthiasis is more likely to occur in those settings, increasing the risk of stunting and diarrhoea (Fauziah et al. 2022). All these results are statistically significant (Table A5).

The primary mechanism by which literature suggests that weather shocks affect child nutrition is via harming the agricultural systems. To explore this mechanism further, table A6 shows the heterogeneity of the weather shock on variables associated with agriculture and farming. It is found that owning a rice field increases the chances of stunting in case of extreme rainfall by 20 percentage points. In contrast, those households whose income comes from the agricultural sector benefited from the shock. A possible explanation is that the timing of the extreme rainfall might have affected rice growing or harvesting but other crops were intact. Therefore, agriculture may not have been the main pathway by which La Niña event impacted stunting levels. Lastly, those owning more livestock (children living in households owning the highest 20th percentile) had lower likelihood of being stunted after the La Niña event compared to those who held the least (all these results are significant). It has been suggested that owning livestock could be related to stunting reduction (Mosites et al. 2015). Additionally, these findings support that owning livestock is a potential coping mechanism for shocks (Ansah, Gardebroek, and Ihle 2021).

The pathway in which CCT reduce stunting is usually associated with increased income and through the CCT conditionalities. Table A7 explores the effect of PKH on different outcomes. CCT increase overall monthly expenditure and spending on protein-rich foods such as milk and egg. However, overall food expenditure did not increase. None of these results are significant and this can be associated with a failure in the design and implementation of the programme, as suggested in the literature (Bastagli et al. 2019).

PKH also increased health-seeking behaviours such as higher complete immunisation by age and higher frequency of weighing in the last three months at the community clinic. These results are consistent with other cash transfer programmes (Bastagli et al. 2019). Despite the conditionality of administering vitamin A to children, the intake of this vitamin was lower in the CCT recipients. Additionally, the PKH programme did not report a decrease in acute illnesses such as diarrhoea, fever, or cough in the month before the survey. However, Cahyadi et al. (2020) argue that chronically ill children would have more low-growth pathways, resulting in stunting, which is not the case.

 

7. Robustness checks

Robustness tests are used in empirical studies to evaluate how much an estimate from different model specifications is supported by estimates from a baseline model (Neumayer and Plümper 2017). Several of these tests are conducted to support the results of this paper.

 

7.1 Variating intensity in rainfall

In theory, it is not very well defined whether more extreme weather rainfall has worse or lesser effects on stunting. It is possible that larger weather shocks have greater effects; for instance, death of people, infrastructure damage and major food security problems (Atanga and Tankpa 2021). However, in case of larger weather shocks, there are a more significant mobilisation of supplies, more extreme coping mechanisms, and in some situations, an intensification of social programmes (Andalón et al. 2016).

This study uses a ZSI value of 2 standard deviations above the yearly-specific mean. New variables using different cut-off ZSI values (1.5 and 2.5 standard deviations) are calculated to investigate the sensitivity of the threshold used for this study. Columns 1 and 2 of table A8 show the full model results using a milder and stronger weather shock. ZSI of 1.5 standard deviations reduces stunting (not significant), and ZSI >2.5 remain substantively similar to the baseline model.

 

7.2 Using a different rainfall variable

Weather shocks can also be defined as cumulative exposure to extreme rainfall instead of an isolated event (Nsabimana and Mensah 2020). Therefore, a new weather shock is created. It counts the years between 2007 and 2013 with a ZSI>2. Column 3 of Table A8 shows the results of the main empirical model. The results remain essentially the same, which suggests that during that time frame, only the La Niña event had such an impact on the population.

 

7.3 Increasing the size of the sample

The population was limited to children under 5 years old because z-scores above that age are constructed on the CDC measure charts. Such charts are based on children in the United States and overestimate the prevalence of stunting (Moelyo, Candrarukmi, and Rachma 2022). Column 4 of Table A8 includes all children in the sample, the age range is 0-16. The effects of weather shocks are almost minimal, but CCT keep proving effective. The interaction term coefficient is fairly similar to the baseline model.

 

7.4 Attrition

This randomised control trial is implemented in 2007 and the last follow-up survey was conducted in 2013. As result, this study might suffer from attrition, i.e. loss of participants throughout the study (Nunan, Aronson, and Bankhead 2018). However, by the last survey, 95.1% of the households surveyed at baseline were still found. This tendency is similar for treatment and control groups (Table 13). The low levels of attrition of the PKH programme suggest that full families do not migrate, and children remain at the same address. As a consequence, they experienced the weather shocks in the subdistrict where they were surveyed (Cahyadi et al. 2020).

 

8. Limitations and scope for further research

This analysis has limitations additional to the data constraints (section 3.2). The primary critical concern is whether this analysis allows for causal inference. The evaluation of the PKH programme on nutritional outcomes uses an instrumental variable to deal with endogeneity, fixed effects to deal with omitted variable bias, and various levels of controls that otherwise might confound the results. This approach supports causal interpretation. However, proving causal effects of rainfall on stunting is not possible. Therefore, the empirical strategy using the interaction of weather shock and CT may not be able to prove causality. However, the exogenous nature of the precipitation increases the robustness of the results.

Additionally, the suggested mechanisms by which CCT mitigate (or not) extreme precipitation shocks are only considerations and not the aim of this study. The literature has suggested various mechanisms using isolated pathways (agriculture, household characteristics, health characteristics, etc). However, strict research exploring those mechanisms must be in the agenda of those studying CCT, weather shocks and public health.

Internal validity in this research is prioritised and achieved by focusing on 1) one country, 2) a very specific population (people living in poverty), and 3) a specific weather shock. However, this might compromise external validity, given that the relationships and conclusions cannot be generalised to other settings (Steckler and McLeroy 2008). Even though CCT are gaining popularity, conditionalities and the amount of cash transferred vary according to the budget and objectives of the delivery institutions.

 

9. Conclusion

As governments work to mitigate and resolve persistently omnipresent child malnutrition, their most vulnerable citizens are increasingly threatened by severe weather shocks due to climate change. Some mitigation strategies are already well recognized and widely implemented, but they are not always accessible for the poorest. Nevertheless, CT are successful social programmes that have been proven to mitigate weather shocks to a certain extent.

This study investigated the effects of the PKH CCT programme in Indonesia on stunting in children under 5 years old, as well as the impact of the 2010-2011 La Niña ENSO cold weather shock on nutritional outcomes. Principally, this research aimed to explore the effects of this CCT programme on stunting in children that had experience this weather shock. Additionally, several mechanisms were explored to understand the pathways by which PKH and weather shocks affect child nutrition according to the UNICEF conceptual framework on child and maternal nutrition.

The analysis reveals that PKH was effective at reducing stunting in all children, particularly girls. Furthermore, the La Niña event correlates with increased stunting particularly in rural areas, mostly girls and children who experienced the event in their first and second year of life. While boys have a higher prevalence of stunting, they are less affected by weather shocks.

When analysing the interaction of the shock and CCT, it was found that PKH did not help mitigate the negative effects on nutrition caused by La Niña event. In fact, PKH is better at reducing stunting levels in the absence of weather shocks, particularly in household with better hygiene resources.

Remarkably, PKH was found effective at -fully- mitigating the effects the La Niña in children whose parents were unemployed or illiterate, in rural areas. These results are not statistically significant and must be interpreted with care. Nevertheless, this emphasises the potential of CCT to act as insurance against risk in those without a stable income.

To the author’s knowledge, only one study has explored the capability of CCT as a mitigation tool for wet weather shocks focusing on nutritional outcomes such as stunting. League and Dylan’s (2022) evaluation of Bolsa Familia in Brazil seems to protect beneficiaries against mild weather shocks in utero. This study contributes to the literature by studying children’s nutritional outcomes in settings outside Latin America, by focussing on a recurrent severe weather event and by not focussing on in utero shocks only.

Further efforts studying how CCT could mitigate extreme rainfall and effects on child nutrition is required to effectively plan weather mitigation programmes or indeed redesign CCT with the purpose of relieving the most vulnerable.

 

REFERENCES

 

Ademas, Ayechew, Metadel Adane, Awoke Keleb, Gete Berihun, and Getu Tesfaw. 2021. ‘Water,
Sanitation, and Hygiene as a Priority Intervention for Stunting in under-Five Children in
Northwest Ethiopia: A Community-Based Cross-Sectional Study’. Italian Journal of Pediatrics
47 (1): 174. https://doi.org/10.1186/s13052-021-01128-y.

Adhvaryu, Achyuta, James Fenske, Namrata Kala, and Anant Nyshadham. 2015. ‘Working Paper.
Fetal Origins of Mental Health: Evidence from Africa’.

Adhvaryu, Achyuta, Anant Nyshadham, Teresa Molina, and Jorge Tamayo. 2018. ‘Helping
Children Catch Up: Early Life Shocks and the PROGRESA Experiment’. w24848.
Cambridge, MA: National Bureau of Economic Research.
https://doi.org/10.3386/w24848.

Aguilar, Arturo, and Marta Vicarelli. 2022. ‘El Niño and Children: Medium-Term Effects of EarlyLife Weather Shocks on Cognitive and Health Outcomes’. World Development 150
(February): 105690. https://doi.org/10.1016/j.worlddev.2021.105690.

Aji, Priasto. 2015. ‘Summary of Indonesia’s Poverty Analysis’. ADB Papers on Indonesia. Asian
Development Bank.

Alatas, Vivi. 2011. ‘Main Findings from the Impact Evaluation of Indonesia’s Pilot Household
Conditional Cash Transfer Progra’. Jakarta: The World Bank.
https://documents.worldbank.org/en/publication/documentsreports/documentdetail/589171468266179965/program-keluarga-harapan-impactevaluation-of-indonesias-pilot-household-conditional-cash-transfer-program.

Amadu, Iddrisu, Abdul-Aziz Seidu, Eric Duku, James Boadu Frimpong, John Elvis Hagan Jnr,
Richard Gyan Aboagye, Belinda Ampah, Collins Adu, and Bright Opoku Ahinkorah. 2021.
‘Risk Factors Associated with the Coexistence of Stunting, Underweight, and Wasting in
Children under 5 from 31 Sub-Saharan African Countries’. BMJ Open 11 (12): e052267.
https://doi.org/10.1136/bmjopen-2021-052267.

Andalón, Mabel, João Pedro Azevedo, Carlos Rodríguez-Castelán, Viviane Sanfelice, and Daniel
Valderrama-González. 2016. ‘Weather Shocks and Health at Birth in Colombia’. World
Development 82 (June): 69–82. https://doi.org/10.1016/j.worlddev.2016.01.015.

Angrist, Joshua D., and Alan B. Krueger. 2001. ‘Instrumental Variables and the Search for
Identification: From Supply and Demand to Natural Experiments’. Journal of Economic
Perspectives 15 (4): 69–85. https://doi.org/10.1257/jep.15.4.69.

Ansah, Isaac Gershon Kodwo, Cornelis Gardebroek, and Rico Ihle. 2021. ‘Shock Interactions,
Coping Strategy Choices and Household Food Security’. Climate and Development 13 (5):
414–26. https://doi.org/10.1080/17565529.2020.1785832.

Asfaw, Solomon, Alessandro Carraro, Benjamin Davis, Sudhanshu Handa, and David Seidenfeld.
2017. ‘Cash Transfer Programmes, Weather Shocks and Household Welfare: Evidence
from a Randomised Experiment in Zambia’. Journal of Development Effectiveness 9 (4): 419–
42. https://doi.org/10.1080/19439342.2017.1377751.

Atanga, Raphael Ane, and Vitus Tankpa. 2021. ‘Climate Change, Flood Disaster Risk and Food
Security Nexus in Northern Ghana’. Frontiers in Sustainable Food Systems 5.
https://www.frontiersin.org/articles/10.3389/fsufs.2021.706721.

Auffhammer, Maximilian, Solomon M. Hsiang, Wolfram Schlenker, and Adam Sobel. 2013. ‘Using
Weather Data and Climate Model Output in Economic Analyses of Climate
Change’. Review of Environmental Economics and Policy 7 (2): 181–98.
https://doi.org/10.1093/reep/ret016.

Bastagli, Francesca, Jessica Hagen-Zanker, Luke Harman, Valentina Barca, Georgina Sturge, and
Tanja Schmidt. 2019. ‘The Impact of Cash Transfers: A Review of the Evidence from
Low- and Middle-Income Countries’. Journal of Social Policy 48 (3): 569–94.
https://doi.org/10.1017/S0047279418000715.

Beal, Ty, Alison Tumilowicz, Aang Sutrisna, Doddy Izwardy, and Lynnette M. Neufeld. 2018. ‘A
Review of Child Stunting Determinants in Indonesia’. Maternal & Child Nutrition 14 (4):
e12617. https://doi.org/10.1111/mcn.12617.

Benhassine, Najy, Florencia Devoto, Esther Duflo, Pascaline Dupas, and Victor Pouliquen. 2015.
‘Turning a Shove into a Nudge? A “Labeled Cash Transfer” for Education’. American
Economic Journal: Economic Policy 7 (3): 86–125. https://doi.org/10.1257/pol.20130225.

Bhavnani, Darlene, Jason E. Goldstick, William Cevallos, Gabriel Trueba, and Joseph N. S.
Eisenberg. 2014. ‘Impact of Rainfall on Diarrheal Disease Risk Associated with
Unimproved Water and Sanitation’. The American Journal of Tropical Medicine and Hygiene 90
(4): 705–11. https://doi.org/10.4269/ajtmh.13-0371.

Black, Robert E., Cesar G. Victora, Susan P. Walker, Zulfiqar A. Bhutta, Parul Christian, Mercedes
de Onis, Majid Ezzati, et al. 2013. ‘Maternal and Child Undernutrition and Overweight in
Low-Income and Middle-Income Countries’. The Lancet 382 (9890): 427–51.
https://doi.org/10.1016/S0140-6736(13)60937-X.

Boening, Carmen, Josh K. Willis, Felix W. Landerer, R. Steven Nerem, and John Fasullo. 2012.
‘The 2011 La Niña: So Strong, the Oceans Fell’. Geophysical Research Letters 39 (19).
https://doi.org/10.1029/2012GL053055.

Brown, Molly E., and Varun Kshirsagar. 2015. ‘Weather and International Price Shocks on Food
Prices in the Developing World’. Global Environmental Change 35 (November): 31–40.
https://doi.org/10.1016/j.gloenvcha.2015.08.003.

Cahyadi, Nur, Rema Hanna, Benjamin A. Olken, Rizal Adi Prima, Elan Satriawan, and Ekki
Syamsulhakim. 2020. ‘Cumulative Impacts of Conditional Cash Transfer Programs:
Experimental Evidence from Indonesia’. American Economic Journal: Economic Policy 12 (4):
88–110. https://doi.org/10.1257/pol.20190245.

Cai, Wenju, Agus Santoso, Guojian Wang, Sang-Wook Yeh, Soon-Il An, Kim M. Cobb, Mat
Collins, et al. 2015. ‘ENSO and Greenhouse Warming’. Nature Climate Change 5 (9): 849–
59. https://doi.org/10.1038/nclimate2743.

Campbell, Lindsay P., Caylor Luther, David Moo-Llanes, Janine M. Ramsey, Rogelio DanisLozano, and A. Townsend Peterson. 2015. ‘Climate Change Influences on Global
Distributions of Dengue and Chikungunya Virus Vectors’. Philosophical Transactions of the
Royal Society B: Biological Sciences 370 (1665): 20140135.
https://doi.org/10.1098/rstb.2014.0135.

Carraro, Alessandro, and Lucia Ferrone. 2020. ‘How Effective Are Cash Transfers in Mitigating
Shocks for Vulnerable Children? Evidence on the Impact of the Lesotho Child Grant
Programme on Multiple Deprivations’, November, 50.

Carrillo, Bladimir. 2020. ‘Early Rainfall Shocks and Later-Life Outcomes: Evidence from
Colombia’. The World Bank Economic Review 34 (1): 179–209.
https://doi.org/10.1093/wber/lhy014.

Colin Cameron, A., and Douglas L. Miller. 2015. ‘A Practitioner’s Guide to Cluster-Robust
Inference’. Journal of Human Resources 50 (2): 317–72. https://doi.org/10.3368/jhr.50.2.317.

Cornwell, Katy, and Brett Inder. 2015. ‘Child Health and Rainfall in Early Life’. The Journal of
Development Studies 51 (7): 865–80. https://doi.org/10.1080/00220388.2014.976618.

Danaei, Goodarz, Kathryn G. Andrews, Christopher R. Sudfeld, Günther Fink, Dana Charles
McCoy, Evan Peet, Ayesha Sania, Mary C. Smith Fawzi, Majid Ezzati, and Wafaie W.
Fawzi. 2016. ‘Risk Factors for Childhood Stunting in 137 Developing Countries: A
Comparative Risk Assessment Analysis at Global, Regional, and Country Levels’. PLOS
Medicine 13 (11): e1002164. https://doi.org/10.1371/journal.pmed.1002164.

Dewey, Kathryn G., and Khadija Begum. 2011. ‘Long-Term Consequences of Stunting in Early
Life’. Maternal & Child Nutrition 7 Suppl 3 (October): 5–18.
https://doi.org/10.1111/j.1740-8709.2011.00349.x.

Dimitrova, Anna. 2021. ‘Seasonal Droughts and the Risk of Childhood Undernutrition in
Ethiopia’. World Development 141 (May): 105417.
https://doi.org/10.1016/j.worlddev.2021.105417.

Dimitrova, Anna, and Raya Muttarak. 2020. ‘After the Floods: Differential Impacts of Rainfall
Anomalies on Child Stunting in India’. Global Environmental Change 64 (September): 102130.
https://doi.org/10.1016/j.gloenvcha.2020.102130.

Drummond, Elizabeth, Fiona Watson, and Jessica Blankenship. 2021. ‘Southeast Asia Regional
Report on Maternal Nutrition and Complementary Feeding’. UNICEF East Asia and
Pacific Region.

Dulkiah, Moh, Leonardo Sari Avid, and Irwandi. 2018. ‘The Impact of Conditional Cash Transfer
(CCT) to Socio-Economic of Poor Families; A Case Study.’ 7 (1): 32–39.

Fauziah, Nisa, Muhammad Abdurrahman Ar-Rizqi, Sofia Hana, Nurul Mufliha Patahuddin, and
Ajib Diptyanusa. 2022. ‘Stunting as a Risk Factor of Soil-Transmitted Helminthiasis in
Children: A Literature Review’. Interdisciplinary Perspectives on Infectious Diseases 2022 (August):
8929025. https://doi.org/10.1155/2022/8929025.

Fitz, DYLAN, and Riley League. 2021. ‘School, Shocks, and Safety Nets: Can Conditional Cash
Transfers Protect Human Capital Investments during Rainfall Shocks?’ The Journal of
Development Studies 57 (12): 2002–26. https://doi.org/10.1080/00220388.2021.1928640.

Fujii, Tomoki, Abu S. Shonchoy, and Sijia Xu. 2016. ‘Impact of Electrification on Children’s
Nutritional Status in Rural Bangladesh’. World Development 102 (February): 315–30.
https://doi.org/10.1016/j.worlddev.2017.07.016.

Gloede, Oliver, Lukas Menkhoff, and Hermann Waibel. 2015. ‘Shocks, Individual Risk Attitude,
and Vulnerability to Poverty among Rural Households in Thailand and Vietnam’. World
Development, Vulnerability to Poverty in South-East Asia: Drivers, Measurement,
Responses, and Policy Issues, 71 (July): 54–78.
https://doi.org/10.1016/j.worlddev.2013.11.005.

Groot, Richard de, Tia Palermo, Sudhanshu Handa, and Luigi Peter Ragno. 2017. ‘Cash Transfers
and Child Nutrition: Pathways and Impacts’. Development Policy Review 35 (5): 621–43.
https://doi.org/10.1111/dpr.12255.

Groot, Richard de, Tia Palermo, Sudhanshu Handa, Luigi Peter Ragno, and Amber Peterman.
2015. ‘Cash Transfers and Child Nutrition: What We Know and What We Need to Know’,
33.

Haile, Kaleab. 2021. ‘Cash Transfers, Negative Rainfall Shocks and Child Welfare in Ethiopia’.
Journal of African Economies, December, ejab029. https://doi.org/10.1093/jae/ejab029.

Hallegatte, Stephane, Adrien Vogt-Schilb, Mook Bangalore, and Julie Rozenberg. 2017.
Unbreakable: Building the Resilience of the Poor in the Face of Natural Disasters. Washington, DC:
World Bank. https://doi.org/10.1596/978-1-4648-1003-9.

Harris, Ian, Timothy J. Osborn, Phil Jones, and David Lister. 2020. ‘Version 4 of the CRU TS
Monthly High-Resolution Gridded Multivariate Climate Dataset’. Scientific Data 7 (1): 109.
https://doi.org/10.1038/s41597-020-0453-3.

Heinrich, Carolyn J., and Matthew T. Knowles. 2020. ‘A Fine Predicament: Conditioning,
Compliance and Consequences in a Labeled Cash Transfer Program’. World Development
129 (May): 104876. https://doi.org/10.1016/j.worlddev.2020.104876.

Hirabayashi, Yukiko, Roobavannan Mahendran, Sujan Koirala, Lisako Konoshima, Dai Yamazaki,
Satoshi Watanabe, Hyungjun Kim, and Shinjiro Kanae. 2013. ‘Global Flood Risk under
Climate Change’. Nature Climate Change 3 (9): 816–21.
https://doi.org/10.1038/nclimate1911.

Inter-Agency Standing Committee (IASC). 2017. The Gender Handbook for Humanitarian Action.
United Nations Entity for Gender Equality and the Empowerment of Women (UN
Women). https://www.unwomen.org/en/digital-library/publications/2017/2/thegender-handbook-for-humanitarian-action.

International Production Assessment Divison. n.d. ‘Indonesia Production’. Accessed 4 May 2022.
https://ipad.fas.usda.gov/countrysummary/default.aspx?id=ID.

Janvry, Alain de, Frederico Finan, Elisabeth Sadoulet, and Renos Vakis. 2006. ‘Can Conditional
Cash Transfer Programs Serve as Safety Nets in Keeping Children at School and from
Working When Exposed to Shocks?’ Journal of Development Economics 79 (2): 349–73.
https://doi.org/10.1016/j.jdeveco.2006.01.013.

Kandpal, Eeshani, Harold Alderman, Jed Friedman, Deon Filmer, Junko Onishi, and Jorge Avalos.
2016. ‘A Conditional Cash Transfer Program in the Philippines Reduces Severe Stunting’.
The Journal of Nutrition 146 (9): 1793–1800. https://doi.org/10.3945/jn.116.233684.

Kousky, Carolyn. 2016. ‘Impacts of Natural Disasters on Children’. Future of Children 26 (1): 73–
92.

Lagarde, Mylene, Andy Haines, and Natasha Palmer. 2009. ‘The Impact of Conditional Cash
Transfers on Health Outcomes and Use of Health Services in Low and Middle Income
Countries’. The Cochrane Database of Systematic Reviews 2009 (4): CD008137.
https://doi.org/10.1002/14651858.CD008137.

League, Riley, and Fitz Dylan. 2022. ‘Early-Life Shocks and Childhood Social Programs: Evidence
of Catch-Up in Brazil. Working Paper.’
https://rileyleague.github.io/files/catchup_leaguefitz.pdf.

Ledlie, N. A., H. Alderman, J. L. Leroy, and L. You. 2018. ‘Rainfall Shocks Are Not Necessarily a
Sensitive Early Indicator of Changes in Wasting Prevalence’. European Journal of Clinical
Nutrition 72 (1): 177–78. https://doi.org/10.1038/ejcn.2017.144.

Lenters, Lindsey, Kerri Wazny, and Zulfiqar A. Bhutta. 2016. ‘Management of Severe and
Moderate Acute Malnutrition in Children’. In Reproductive, Maternal, Newborn, and Child
Health: Disease Control Priorities, Third Edition (Volume 2), edited by Robert E. Black, Ramanan
Laxminarayan, Marleen Temmerman, and Neff Walker. Washington (DC): The
International Bank for Reconstruction and Development / The World Bank.
http://www.ncbi.nlm.nih.gov/books/NBK361900/.

Leroy, Jef L, and Edward A Frongillo. 2019. ‘Perspective: What Does Stunting Really Mean? A
Critical Review of the Evidence’. Advances in Nutrition 10 (2): 196–204.
https://doi.org/10.1093/advances/nmy101.

Leroy, Jef L., Marie Ruel, and Ellen Verhofstadt. 2009. ‘The Impact of Conditional Cash Transfer
Programmes on Child Nutrition: A Review of Evidence Using a Programme Theory
Framework’. Journal of Development Effectiveness 1 (2): 103–29.
https://doi.org/10.1080/19439340902924043.

Levine, David, and Dean Yang. 2014. ‘The Impact of Rainfall on Rice Output in Indonesia’.
w20302. Cambridge, MA: National Bureau of Economic Research.
https://doi.org/10.3386/w20302.

Li, Xianghu, Qi Zhang, and Xuchun Ye. 2013. ‘Dry/Wet Conditions Monitoring Based on TRMM
Rainfall Data and Its Reliability Validation over Poyang Lake Basin, China’. Water 5 (4):
1848–64. https://doi.org/10.3390/w5041848.

Li, Yan, Kaiyu Guan, Gary D. Schnitkey, Evan DeLucia, and Bin Peng. 2019. ‘Excessive Rainfall
Leads to Maize Yield Loss of a Comparable Magnitude to Extreme Drought in the United
States’. Global Change Biology 25 (7): 2325–37. https://doi.org/10.1111/gcb.14628.

Lieber, Mark, Peter Chin-Hong, Knox Kelly, Madhavi Dandu, and Sheri D. Weiser. 2022. ‘A
Systematic Review and Meta-Analysis Assessing the Impact of Droughts, Flooding, and
Climate Variability on Malnutrition’. Global Public Health 17 (1): 68–82.
https://doi.org/10.1080/17441692.2020.1860247.

Maccini, Sharon, and Dean Yang. 2009. ‘Under the Weather: Health, Schooling, and Economic
Consequences of Early-Life Rainfall’. American Economic Review 99 (3): 1006–26.
https://doi.org/10.1257/aer.99.3.1006.

Manley, James, Harold Alderman, and Ugo Gentilini. 2022. ‘More Evidence on Cash Transfers
and Child Nutritional Outcomes: A Systematic Review and Meta-Analysis’. BMJ Global
Health 7 (4): e008233. https://doi.org/10.1136/bmjgh-2021-008233.

Manley, James, Seth Gitter, and Vanya Slavchevska. 2012. ‘How Effective Are Cash Transfer
Programmes at Improving Nutritional Status? A Rapid Evidence Assessment of
Programmes’ Effects on Anthropometric Outcomes’, July, 94.

Márquez-González, Horacio, and Verónica Marlene García-Sámano. 2012. ‘Clasificación y
evaluación de la desnutrición en el paciente pediátrico’ 7: 11.

McGuire, Joel, Caspar Kaiser, and Anders M. Bach-Mortensen. 2022. ‘A Systematic Review and
Meta-Analysis of the Impact of Cash Transfers on Subjective Well-Being and Mental
Health in Low- and Middle-Income Countries’. Nature Human Behaviour 6 (3): 359–70.
https://doi.org/10.1038/s41562-021-01252-z.

Mehta, Nilesh M., Mark R. Corkins, Beth Lyman, Ainsley Malone, Praveen S. Goday, Liesje
Nieman Carney, Jessica L. Monczka, Steven W. Plogsted, W. Frederick Schwenk, and
American Society for Parenteral and Enteral Nutrition Board of Directors. 2013. ‘Defining
Pediatric Malnutrition: A Paradigm Shift toward Etiology-Related Definitions’. JPEN.
Journal of Parenteral and Enteral Nutrition 37 (4): 460–81.
https://doi.org/10.1177/0148607113479972.

Moelyo, Annang Giri, Dewinda Candrarukmi, and Ulfa Puspita Rachma. 2022. ‘Using the National
Indonesian Growth Chart to Assess Short Stature and Obesity in Urban Schoolchildren in
Surakarta, Indonesia: Comparisons to the WHO 2007 and CDC 2000 Growth Charts’.
Paediatrica Indonesiana 62 (3): 180–85. https://doi.org/10.14238/pi62.3.2022.180-5.

Mohammadi-Nasrabadi, Fatemeh. 2016. ‘Impact of Cash Transfer on Food Security: A Review’.
Nutrition and Food Sciences Research 3 (2): 3–10.
https://doi.org/10.18869/acadpub.nfsr.3.2.3.

Mosites, Emily M., Peter M. Rabinowitz, Samuel M. Thumbi, Joel M. Montgomery, Guy H.
Palmer, Susanne May, Ali Rowhani-Rahbar, Marian L. Neuhouser, and Judd L. Walson.
2015. ‘The Relationship between Livestock Ownership and Child Stunting in Three
Countries in Eastern Africa Using National Survey Data’. PLoS ONE 10 (9): e0136686.
https://doi.org/10.1371/journal.pone.0136686.

Naylor, Rosamond L., David S. Battisti, Daniel J. Vimont, Walter P. Falcon, and Marshall B. Burke.
2007. ‘Assessing Risks of Climate Variability and Climate Change for Indonesian Rice
Agriculture’. Proceedings of the National Academy of Sciences 104 (19): 7752–57.
https://doi.org/10.1073/pnas.0701825104.

Neumayer, Eric, and Thomas Plümper. 2017. Robustness Tests for Quantitative Research. 1st ed.
Cambridge University Press. https://doi.org/10.1017/9781108233590.

Newhouse, J. P., and M. McClellan. 1998. ‘Econometrics in Outcomes Research: The Use of
Instrumental Variables’. Annual Review of Public Health 19: 17–34.
https://doi.org/10.1146/annurev.publhealth.19.1.17.

Noor, NA Mohd, Nai Mohd Noor, R Alias, and M Mohamad Ideris. 2020. ‘Drought Indices
Monitoring Using SPI and Z Index Score for Gua Musang, Kelantan’. IOP Conference Series:
Materials Science and Engineering 932 (1): 012050. https://doi.org/10.1088/1757-
899X/932/1/012050.

Nsabimana, Aimable, and Justice Tei Mensah. 2020. ‘Weather Shocks and Child Nutrition:
Evidence from Tanzania’, May. https://doi.org/doi:10.35188/UNU-WIDER/2020/814-
6.

Nunan, David, Jeffrey Aronson, and Clare Bankhead. 2018. ‘Catalogue of Bias: Attrition Bias’.
BMJ Evidence-Based Medicine 23 (1): 21–22. https://doi.org/10.1136/ebmed-2017-110883.

Olofin, Ibironke, Christine M. McDonald, Majid Ezzati, Seth Flaxman, Robert E. Black, Wafaie
W. Fawzi, Laura E. Caulfield, Goodarz Danaei, and Nutrition Impact Model Study
(anthropometry cohort pooling). 2013. ‘Associations of Suboptimal Growth with AllCause and Cause-Specific Mortality in Children under Five Years: A Pooled Analysis of
Ten Prospective Studies’. PloS One 8 (5): e64636.
https://doi.org/10.1371/journal.pone.0064636.

Onis, Mercedes de, and Francesco Branca. 2016. ‘Childhood Stunting: A Global Perspective’.
Maternal & Child Nutrition 12 Suppl 1 (May): 12–26. https://doi.org/10.1111/mcn.12231.

Özaltin, Emre, Kenneth Hill, and S. V. Subramanian. 2010. ‘Association of Maternal Stature With
Offspring Mortality, Underweight, and Stunting in Low- to Middle-Income Countries’.
JAMA : The Journal of the American Medical Association 303 (15): 1507–16.
https://doi.org/10.1001/jama.2010.450.

Phalkey, Revati K., Clara Aranda-Jan, Sabrina Marx, Bernhard Höfle, and Rainer Sauerborn. 2015.
‘Systematic Review of Current Efforts to Quantify the Impacts of Climate Change on
Undernutrition’. Proceedings of the National Academy of Sciences of the United States of America 112
(33): E4522–29. https://doi.org/10.1073/pnas.1409769112.

Premand, Patrick, and Quentin Stoeffler. 2020. ‘Do Cash Transfers Foster Resilience? Evidence
from Rural Niger’. Working Paper. Washington, DC: World Bank.
https://doi.org/10.1596/1813-9450-9473.

Rabassa, Mariano, Emmanuel Skoufias, and Hanan Jacoby. 2014. ‘Weather and Child Health in
Rural Nigeria’. Journal of African Economies 23 (4): 464–92.
https://doi.org/10.1093/jae/eju005.

Rodysill, Jessica R., James M. Russell, Mathias Vuille, Sylvia Dee, Brent Lunghino, and Satria
Bijaksana. 2019. ‘La Niña-Driven Flooding in the Indo-Pacific Warm Pool during the Past
Millennium’. Quaternary Science Reviews 225 (December): 106020.
https://doi.org/10.1016/j.quascirev.2019.106020.

Rosinger, Asher Y. 2018. ‘Household Water Insecurity after a Historic Flood: Diarrhea and
Dehydration in the Bolivian Amazon’. Social Science & Medicine 197 (January): 192–202.
https://doi.org/10.1016/j.socscimed.2017.12.016.

Ruel, Marie T. 2010. ‘The Oriente Study: Program and Policy Impacts’. The Journal of Nutrition 140
(2): 415–18. https://doi.org/10.3945/jn.109.114512.

Ruel, Marie T., and Harold Alderman. 2013. ‘Nutrition-Sensitive Interventions and Programmes:
How Can They Help to Accelerate Progress in Improving Maternal and Child Nutrition?’
The Lancet 382 (9891): 536–51. https://doi.org/10.1016/S0140-6736(13)60843-0.

Ruggles, Steven, Steven M. Manson, Tracy A. Kugler, David A. Haynes II, David C. Van Riper,
and Maryia Bakhtsiyarava. 2018. ‘IPUMS Terra: Integrated Data on Population and
Environment: Version 2’. Minneapolis, MN: IPUMS.
https://doi.org/10.18128/D090.V2.

Schäfer, Laura, Eleanor Waters, Sönke Kreft, Michael Zissener, Peter Hoeppe, Christoph Bals,
Simone Ruiz-Vergote, et al. 2016. ‘Making Climate Risk Insurance Work for the Most
Vulnerable: Seven Guiding Principles’. Policy Report. UNU-EHS Publication. Munich
Climate Insurance Initiative (MCII).

Soekatri, Moesijanti Y. E., Sandjaja Sandjaja, and Ahmad Syauqy. 2020. ‘Stunting Was Associated
with Reported Morbidity, Parental Education and Socioeconomic Status in 0.5–12-YearOld Indonesian Children’. International Journal of Environmental Research and Public Health 17
(17): 6204. https://doi.org/10.3390/ijerph17176204.

Soliman, Ashraf, Vincenzo De Sanctis, Nada Alaaraj, Shayma Ahmed, Fawziya Alyafei, Noor
Hamed, and Nada Soliman. 2021. ‘Early and Long-Term Consequences of Nutritional
Stunting: From Childhood to Adulthood’. Acta Bio Medica : Atenei Parmensis 92 (1):
e2021168. https://doi.org/10.23750/abm.v92i1.11346.

Sparrow, Robert. 2008. ‘Conditional Cash Transfers in Indonesia: Baseline Survey Report.
Program Keluarga Harapan and PNPM-Generasi’. The World Bank.

Steckler, Allan, and Kenneth R. McLeroy. 2008. ‘The Importance of External Validity’. American
Journal of Public Health 98 (1): 9–10. https://doi.org/10.2105/AJPH.2007.126847.

Tang, Youmin, Rong-Hua Zhang, Ting Liu, Wansuo Duan, Dejian Yang, Fei Zheng, Hongli Ren,
et al. 2018. ‘Progress in ENSO Prediction and Predictability Study’. National Science Review
5 (6): 826–39. https://doi.org/10.1093/nsr/nwy105.

The World Bank. 2021. ‘World Bank Climate Change Knowledge Portal’. 2021.
https://climateknowledgeportal.worldbank.org/.

Thiede, Brian C., and Clark Gray. 2020. ‘Climate Exposures and Child Undernutrition: Evidence
from Indonesia’. Social Science & Medicine 265 (November): 113298.
https://doi.org/10.1016/j.socscimed.2020.113298.

Thurstans, Susan, Charles Opondo, Andrew Seal, Jonathan Wells, Tanya Khara, Carmel Dolan,
André Briend, et al. 2020. ‘Boys Are More Likely to Be Undernourished than Girls: A
Systematic Review and Meta-Analysis of Sex Differences in Undernutrition’. BMJ Global
Health 5 (12): e004030. https://doi.org/10.1136/bmjgh-2020-004030.

Tirado, M. C., P. Crahay, L. Mahy, C. Zanev, M. Neira, S. Msangi, R. Brown, C. Scaramella, D.
Costa Coitinho, and A. Müller. 2013. ‘Climate Change and Nutrition: Creating a Climate
for Nutrition Security’. Food and Nutrition Bulletin 34 (4): 533–47.
https://doi.org/10.1177/156482651303400415.

Torlesse, Harriet, Aidan Anthony Cronin, Susy Katikana Sebayang, and Robin Nandy. 2016.
‘Determinants of Stunting in Indonesian Children: Evidence from a Cross-Sectional
Survey Indicate a Prominent Role for the Water, Sanitation and Hygiene Sector in Stunting
Reduction’. BMC Public Health 16 (1): 669. https://doi.org/10.1186/s12889-016-3339-8.

UNICEF. 2007. ‘Progress for Children.A Wolrd Fit for Children Statistical Review’, December.

United Nations Children’s Fund (UNICEF). 2021. ‘UNICEF Conceptual Framework on Maternal
and Child Nutrition’. UNICEF Nutrition and Child Development Section, Programme
Group. https://www.unicef.org/documents/conceptual-framework-nutrition.

Utami, Ressa Andriyanu, Agus Setiawan, and Poppy Fitriyani. 2019. ‘Identifying causal risk factors
for stunting in children under five years of age in South Jakarta, Indonesia’. Enfermería
Clínica 29 (September): 606–11. https://doi.org/10.1016/j.enfcli.2019.04.093.

Väänänen, Elina, Katharina Nett, Cecilia Costella, and Janot Mendler de Suarez. 2019. ‘Linking
Climate Risk Insurance with Shock-Responsive Social Protection’. InsuResilience Global
Partnership.

Vaivada, Tyler, Nadia Akseer, Selai Akseer, Ahalya Somaskandan, Marianne Stefopulos, and
Zulfiqar A Bhutta. 2020. ‘Stunting in Childhood: An Overview of Global Burden, Trends,
Determinants, and Drivers of Decline’. The American Journal of Clinical Nutrition 112
(Supplement_2): 777S-791S. https://doi.org/10.1093/ajcn/nqaa159.

Vicarelli, Marta. 2011. ‘Exogenous Income Shocks and Consumption Smoothing Strategies
Among Rural Households in Mexico’. Ph. D. Thesis, Columbia University.
https://www.academia.edu/1328691/Exogenous_Income_Shocks_and_Consumption_
Smoothing_Strategies_Among_Rural_Households_in_Mexico.

Victora, Cesar Gomes, Mercedes de Onis, Pedro Curi Hallal, Monika Blössner, and Roger
Shrimpton. 2010. ‘Worldwide Timing of Growth Faltering: Revisiting Implications for
Interventions’. Pediatrics 125 (3): e473-480. https://doi.org/10.1542/peds.2009-1519.

Wamani, Henry, Anne Nordrehaug Åstrøm, Stefan Peterson, James K. Tumwine, and Thorkild
Tylleskär. 2007. ‘Boys Are More Stunted than Girls in Sub-Saharan Africa: A MetaAnalysis of 16 Demographic and Health Surveys’. BMC Pediatrics 7 (1): 17.
https://doi.org/10.1186/1471-2431-7-17.

Watts, Nick, Markus Amann, Sonja Ayeb-Karlsson, Kristine Belesova, Timothy Bouley, Maxwell
Boykoff, Peter Byass, et al. 2018. ‘The Lancet Countdown on Health and Climate Change:
From 25 Years of Inaction to a Global Transformation for Public Health’. The Lancet 391
(10120): 581–630. https://doi.org/10.1016/S0140-6736(17)32464-9.

Wisana, I Dewa Gede Karma. 2015. ‘Rural Households’ Characteristics on Consumption
Vulnerability Due to Natural Disasters: The Case of Floods and Earthquakes in Indonesia’.
In . Conference Paper.

Wood, Rachel Godfrey. 2011. ‘Is There a Role for Cash Transfers in Climate Change Adaptation?’
IDS Bulletin 42 (6): 79–85. https://doi.org/10.1111/j.1759-5436.2011.00277.x.

World Health Organization. 2006. ‘Growth Standards: Methods and Development’. WHO. World
Health Organization. 2006.
https://www.who.int/childgrowth/standards/technical_report/en/.
———. 2015. ‘Stunting in a Nutshell’. 2015. https://www.who.int/news/item/19-11-2015-
stunting-in-a-nutshell.
———. 2018. Reducing Stunting in Children: Equity Considerations for Achieving the Global Nutrition
Targets 2025. Geneva: World Health Organization.
https://apps.who.int/iris/handle/10665/260202.
———. 2021. ‘Fact Sheets – Malnutrition’. 9 June 2021. https://www.who.int/news-room/factsheets/detail/malnutrition.
———. 2022. ‘Child Growth Standards – Software’. 2022. https://www.who.int/tools/childgrowth-standards/software.

 

 

APPENDICES