THE EFFECTS OF SOCIAL ASSISTANCE PROGRAMS ON STUNTING PREVALENCE RATES IN INDONESIA

Background: Social assistance, such as Non-Cash Food Assistance (BPNT) and the Family Hope Program (PKH), are among the instruments used to eradicate stunting in the short and long term. Therefore, it is important to ensure its effective use by individuals. Aims: This study aimed to review the effect of social assistance on stunting prevalence rates in Indonesia. Methods: This research employs a quantitative approach, utilizing numerical measurements and statistical analysis. The data is secondary and sourced from the Central Bureau of Statistics and the Ministry of Finance. The specific method applied is the generalized method of moments (GMM) dynamic panel regression. The research sample encompasses panel data from 34 provinces from 2015 to 2021. This study utilizes the indicator of very short height for children less than five years of age as a proxy for stunting. Furthermore, low height in children is adopted as a measure of stunting. Furthermore, this study adopted a measure of low height in children to indicate stunting. Results: The results showed that social assistance negatively affected stunting, supported by various other indicators such as the proportion of proper sanitation, the average length of schooling, protein consumption, and drinking water sources. Meanwhile, excessive calorie consumption increased stunting in a certain period. Conclusion: In this condition, government social assistance was urgently needed since access to the lowest decile households increased stunting rates. The practical implications for policy derived from these findings involve optimizing the efficiency of social assistance initiatives, emphasizing the significance of addressing environmental factors in programs aimed at preventing stunting, integrating health and nutrition initiatives, and implementing educational campaigns within communities to raise awareness about calorie consumption.


Introduction
Parents are expected to anticipate exceptional growth and development of their children.However, the expectation is not achieved due to malnutrition-related challenges.In a study conducted by the Ministry of Health, 10.2% of Indonesian children weighed less than 2500 grams, 19.6% exhibited abnormal weight indicative of malnourishment, while 32.2% experienced stunting (Ministry of Health, 2018).The developmental delays were experienced due to malnutrition, and the impact resulted in stunting.According to the World Health Organization (WHO) definition, stunting is manifested when the height of children is minus two standard deviations of the growth median for their age (WHO, 2022).Consequently, stunting children are shorter than their peers of the same age group.
The Indonesian Nutrition Status Survey (SSGI) showed that the prevalence of stunting decreased from 24.4% in 2021 to 21.6% in 2022 (Ministry of Health RI, 2023).Even though a decrease was experienced in 2022, the condition remained a severe problem for children health.Wardani et al. (2022a) stated that stunting causes a decline in cognitive, language, and motor functions as well as several degenerative diseases.This statement was supported by President Joko Widodo at the BKKBN National Work Meeting, where the condition was reported to be dangerous for children's development, leading to poor learning abilities, mental retardation, and chronic diseases (Bureau of Communication and Public Service, 2023).Tampy et al. (2020) stated that normal children could improve their cognitive abilities.Therefore, the condition can affect the quality of human resources in the future.
The causes of stunting are very complex and previous studies showed that there was the influence of maternal factors such as height, age during pregnancy, duration of breastfeeding, dietary nutrition during pregnancy, and maternal education level on the increased risk (Beal et al., 2018;Nurbiah, Rosidi and Margawati, 2019;Tyas and Setyonaluri, 2022;Wardani, Nurrochmah and Mawarni, 2022a).Furthermore, stunted mothers with low education increase the likelihood of stunting (Widyaningsih et al., 2022).A large number of children living at home under age five and the lack of mothers attending antenatal care services can also increase the risk (Titaley et al., 2019).
In another study, Kandpal et al. (2016) examined the Conditional Cash Transfer (CCT) program, one of which was Pantawid for stunting in the Philippines.The program had a significant effect on reducing stunting by using cross-sectional data and providing cash transfers to poor households for education, children's health, and maternal health services.Meanwhile, Muhtar et al. (2022) showed that social assistance programs, as measured through the Family Hope Program (PKH) and the Staple Food Program, reduced stunting in Blora, Central Java.
In several previous studies, stunting has often focused on socioeconomic factors and household characteristics (Akbar, Mahardhika and Sihaloho, 2021), as well as maternal (Beal et al., 2018;Nurbiah et al., 2019;Tyas and Setyonaluri, 2022;Wardani et al., 2022b; ) and environmental factors (Irianti et al., 2019;Cameron et al., 2021;Nizaruddin and Ilham, 2022).Furthermore, there are studies linking stunting to food prices (Ilman and Wibisono, 2019).While only a few studies have directly linked cases of stunting to cash assistance programs like Indonesia's PKH, the program does contain elements that can potentially contribute to addressing stunting.
Though not specifically designed to tackle stunting, PKH provides financial assistance to households, which can improve their access to nutritious food and essential health services.Therefore, the impact of direct assistance on stunting in Indonesia was examined.This study contributes to the reduction of stunting and provides an overview of the importance of direct cash assistance to improve children's nutritional status, impacting health and decreasing the condition.

Method
In this study, a quantitative method was used with a stochastic model to consider the existence of variables with a probability distribution (Gujarati, 2004)

Original Research
Ministry of Finance obtained data on local government spending within the social sector, a crucial component of the relevant social security landscape.The starting point of 2015 aligns with the national trend in stunting reduction, while 2021 is the most recent data available across all provinces.Detailed variable definitions are provided in Table 1.
Based on previous study, social assistance programs negatively affect stunting (Aizawa, 2020;Jibril, Puspitarini and Nawangsih, 2022).Programs such as PKH play an essential role in improving the nutritional status of children and can indirectly reduce stunting rates.In addition, sanitary conditions are an essential part of reducing this condition.Several studies showed that proper sanitation was negatively related to the condition (Rah et al., 2020;Gizaw et al., 2022).Children with proper sanitation are 29% less likely to experience the problem than those with poor sanitation (Rah et al., 2020).Meanwhile, the educational level of parents negatively correlates with stunting (Casale, Espi and Norris, 2018;Chowdhury et al., 2022).Insufficient food consumption has a positive relationship with the prevalence (Wardani, Wulandar and Suharmanto, 2020).In the severe category, stunting children are mostly from families with insufficient food consumption or insecurity.
Nutritional status, measured by calories and protein, correlates with stunting.Caloric intake has a negative correlation (Logarajan et al., 2023) and can reduce the prevalence.However, some studies stated that the proportion of calories from carbohydrates and protein was not a predictor of the condition (Pratiwi, Irawan and Hidayat, 2019).Protein intake negatively correlates with stunting (Ernalia et al., 2018;Rizal, Haya and Maigoda, 2022) and can reduce the incidence.Nutrition, particularly in the form of protein, plays an important role in supporting the growth of children.The absence or deficiency of this essential substance can significantly hinder their development (Rahayu et al., 2020).Other variables, such as internet access, negatively affect stunting (Huo et al., 2022).Accessing the internet facilitates the acquisition of related information, prevention, and effective strategies for overcoming the condition (Talib et al., 2021).The source of drinking water also affects the prevalence.Households with access to proper drinking water sources can reduce stunting (Gupta and Santhya, 2020).Children with an adequate source of drinking water are more likely to prevent the condition than those with access to inadequate drinking water (Mzumara et al., 2018;Sunardi et al., 2021).The effect of social assistance programs can be estimated using dynamic panel regression in the form of GMM.In this study, GMM with a variable lag term was constructed to analyze the impulse response to changes in social assistance units.The transmission of the Arrelanobond and Sargan post-estimator tests is better and more accessible from the classic problems of autocorrelation and heteroscedasticity (Arellano and Bond, 1991;Arellano and Bover, 1995)

Descriptive Statistics
Table 2 shows descriptive statistics and stunting data has an average of 10,963, where the prevalence rate is not too high.The natural logarithm of social assistance also spreads, making it more accurate with the average value.Furthermore, 70,839 households have proper sanitation, and the average length of schooling as measured from years spent by residents aged 15 years and above is nine years or in the Junior High School category.The average insufficient food consumption is 10.785%, which means there are still people who consume food, but slightly less than their energy needs.The nutritional condition as measured by consuming calories and protein has also been shown sufficiently.There is more internet access in cities but the feasibility of drinking water sources in villages is far greater than in cities.

Estimation Results
Table 3 presents the results of regression estimates and focuses on social assistance.The coefficients show that the prevalence of previous stunting has a positive effect on the current result.Specifically, a 1% increase in the previous stunting prevalence leads to an estimated 0.097% increase in the current stunting prevalence.Meanwhile, social assistance consistently has a negative and significant correlation with the prevalence at lag1 and lag2.This suggests that a 1% increase in social assistance in the past year (lag1) is associated with a 0.256% decrease in the current stunting prevalence.Furthermore, the presence of social assistance even one period earlier (lag2) is also consistently associated with a decrease in the current prevalence of stunting.The percentage of households with proper sanitation positively affects the prevalence.However, the proportion of households with proper sanitation is negatively correlated and relatively consistent by adding or subtracting other variables.These results indicate that the higher the proportion of households with proper sanitation, the lower the prevalence of stunting.As seen in the first estimate, when the proportion of proper sanitation increases by 1%, it reduces the prevalence of stunting by 1.08%.Level of education also has a negative and significant effect on the occurrence of stunting.It means the higher the education level, the lower the stunting prevalence rate.In the first estimate, increasing education by one year can reduce the stunting prevalence rate by 5.912%.Insufficient food consumption seen from equations 1 and 6 causes stunting.In the first estimate, when the prevalence of undernourishment increases by 1%, it can increase the prevalence of stunting by 0.131%.Calorie and protein consumption have a positive and negative correlation with the variable.Finally, the percentage of access to the internet and a source of proper drinking water in the village is positively correlated with the prevalence.This suggests that while both internet access and a source of proper drinking water are available within the village, there is still room for improvement in their accessibility or quality.

Discussion
According to Minister of Finance Regulation No 254/PMK.05/2015,Social Assistance Expenditures at State Ministries/ Institutions are in the form of transfers of money, goods, or services provided by the Government to protect society from possible social risks, increase economic capacity, and community welfare.In this context, the concept plays an important role in mitigating social vulnerability, particularly in addressing issues such as stunting.This enables individuals to lead healthy, intelligent lives and give birth to the next generation, free from the burden of stunting.
Social assistance provided is in the form of material and non-material (Ministry of Social Affairs, 2021).Material assistance is usually through programs such as the Smart Indonesia Program (KIP), the National Health Insurance Program (JKN), PKH, and Non-Cash Food Assistance (BPNT).
Meanwhile, non-material assistance is provided through psychoeducation at the Family Development Session (FDS).This form of material and non-material assistance can Munawaroh, Fajri, Ajija Indonesian Journal of Health Administration (Jurnal Administrasi Kesehatan Indonesia) p-ISSN 2303-3592, e-ISSN 2540-9301 10.20473/jaki.v12i1.2024.74-85 .74-85Original Research be used as well as possible by the community following its designation.
The estimation results have a negative correlation with the prevalence of stunting.The results are in line with the study by Kandpal et al. (2016), Aizawa (2020), andJibril et al. (2022), where the existence of social assistance programs reduces the prevalence.Furthermore, the program can be used for spending on children needs by fulfilling nutritious food, which will have a good impact on their growth and development.It is undeniable that social assistance is not used according to children's needs but for other needs.Therefore, socializing and educating the public regarding the importance of maintaining children care and nutrition is important.
The percentage proportion of households with access to proper sanitation services also negatively correlates with the prevalence of stunting.This follows the results of study by Rah et al. (2020) and Gizaw et al. (2022), where it was found that proper sanitation conditions reduce the prevalence.Sanitation factors include the use and ownership of latrines, defecation behavior, and disposal of children's feces in latrines.Furthermore, open defecation can cause developmental delays in children (Bagcchi, 2015) and this behavior contaminates the environment due to the spread of bacteria in feces.Several bacteria can enter the intestines when children touch bacteria and suck their fingers, leading to potential consequences such as appetite disruption, and malnutrition, contributing to the prevalence.
A high level of parental education also reduces the prevalence of stunting.This is in line with Casale et al. (2018) and Chowdhury et al. (2022), where parents' education level negatively correlates with the condition.Parents find it easier and faster to access health information, impacting the growth and development of children with higher education.Even the level of a mother education greatly affects the health of children.This is related to the significant role of forming the habit of eating nutritious food.Therefore, the level of education is closely related to nutritional knowledge in children.
The prevalence of stunting is also associated with insufficient food consumption.Insufficient food consumption has a positive correlation with the prevalence and, according to Wardani et al. (2020), stunting children tend to come from families with insufficient food intake or insecurity.Inadequate food consumption affects the nutritional adequacy of families, specifically for mothers and children.Lack of nutrition negatively impacts children's growth and leads to stunting.
The estimation results showed that calorie consumption had a positive correlation, but in the correlation matrix, calorie and protein consumption negatively impacted stunting.According to Ernalia et al. (2018), Logarajan et al. (2023), andRizal, Haya, andMaigoda (2022), the better the nutrition of children as measured by the consumption of sufficient calories and protein, the less likely there will be a risk of stunting.Protein is needed to develop every cell in the body and maintain immunity (Rizal, Haya and Maigoda, 2022).Calorie intake containing carbohydrates is also the most important energy source for children to experience good growth and be free from malnutrition problems.
Access to the internet and a source of proper drinking water in the village is positively correlated with the prevalence of stunting.This is in contrast to studies by Mzumara et al. (2018), Gupta and Santhya (2020), Sunardi et al. (2021), andTalib et al. (2021), where internet access and adequate drinking water sources are found to reduce the condition.Internet access reflects the possibility that parents have not fully used available resources.In addition, they are not using the access to seek information or knowledge related to the condition but for other purposes.This limited use may not yield significant results in the prevention of the condition.Similarly, proper drinking water sources in the village positively correlate with stunting.The category of having access to a reliable source of drinking water may not necessarily reflect the actual condition for all activities, including drinking, cooking, and toileting.Some households in rural areas still use inadequate water sources for bathing and washing (Nurhidayati and

Original Research
Riyadi, 2022).These efforts have also not yielded significant results in terms of improving access to safe drinking water sources within the village and reducing the prevalence of stunting.This study also had various limitations, for example, the potential for proper sanitation was expanded directly.Furthermore, internet access, which should open up knowledge and information, had a positive impact on increasing stunting.Future research needs to simulate some of the methods used.The positive association between internet access and increased stunting requires further analysis.Research can examine the internet content accessed, the extent to which the information obtained is helpful or even counterproductive to children's health, and identify factors that can increase the effectiveness of sanitation in reducing stunting.

Conclusion
In conclusion, this study was conducted to analyze social assistance programs on the prevalence of stunting in Indonesia.Based on the estimation results, direct social assistance from the government reduced the potential within a certain period.Adequate sanitation and proportions had positive and negative effects on stunting.Meanwhile, the average length of schooling, protein consumption, urban internet access, and rural drinking water sources negatively affected stunting.
Insufficient food consumption, calorie consumption, and drinking water sources in urban areas positively affected stunting.This indicated that the higher assistance given to certain decile people also promoted an increase in decent consumption.However, the increase required specific information from internet accessibility and the availability of food sources as well as water in the surrounding environment.The role of the highly educated head of household promoted knowledge and provided insight into healthy consumption.This potential also depended on the good quality of the surrounding environment to reduce the prevalence.
The implications of this study highlight the importance of various factors in reducing stunting.Comprehensive regulatory efforts are essential to reduce the prevalence of stunting significantly.If used effectively by families, social cash transfers can significantly reduce stunting by helping them meet their children's nutritional and health needs.Through this program, beneficiary families receive cash to buy more nutritious food and access essential health services such as immunizations and regular check-ups.In the long run, cash transfers can improve families' overall social and economic conditions, leading to increased access to essential resources for improving child health.
This study's novelty lies not only in highlighting the significant role of social cash transfers in indirectly reducing stunting but also in linking it to various factors affecting family well-being.These factors include socioeconomic status, health (encompassing nutrition and water sources), and access to information, including the internet.By examining these interconnected aspects, this research provides a valuable foundation for policymakers and relevant agencies to design more effective stunting reduction policies that address the multidimensional nature of the problem.

Table 1 .
Definition of Study Variables

Table 3 .
Estimation Results of the Effect of Social Assistance on Stunting Prevalence , and the empirical model is shown as follows.