Child Stunting and Economic Outcomes in SAARC Countries: The Empirical Evidence

Background: Stunting and its economic consequences are gaining increased awareness. As healthy human capital is the key to higher economic growth in the country, according to UNICEF’s report in 2018, almost 40% of the total stunted population lives in this South Asian region. Because of the long-term implications, this is of great concern to local and international health organizations and agencies. Method: This study explores the causal relationship between the socio-economic determinants of child stunting prevalence under 5 years of age in SAARC countries by constructing fixed-effect modelling and the two-stage least square (2SLS) model using instrumental variables; urbanization, governance stability index, and rainfall and temperature anomalies with the GDP per capita variable from 1984 to 2018. Results: The results reveal that both the variables have a significant causal influence on each other. A 10% rise in gross domestic product per capita reduces child stunting to 6%, implying that economic expansion in this region is presumptively pro-poor. A 1% increase in childhood stunting results in a 3.4% drop in the region's current GDP per capita. The results lie under the critical bound at a 1% level of significance. Conclusion: The study urges the governments of the SAARC countries to adopt pro-poor policies with an effective mechanism for economic growth transition in the targeted area.


2-REVIEW OF LITERATURE
This section reviews selected literature concerning theoretical discussion on the causal relationship between child stunting and economic growth, along with reviewing some empirical studies to explore the dynamics and determinants of both the variables. The review will explore different determinants of stunting, which will help us understand child stunting and its consequences, which will further understand the impact it has on household income and ultimately have an impact on the country's economic growth.

Short term consequences of stunting on health development
• Stunted mothers are a direct risk factor for complications during pregnancy and delivery, as well as the baby's survival (5,20) • Poor nutrition and frequent infections cause a child's nutritional condition to deteriorate and increase vulnerability to infections and diseases (21,22). Infections provide a clear explanation for the child's nutritional state. It causes a decrease in appetite and energy.
• It plummets the absorption of nutrients (6,23). The available nutrients in the child's body are diverted from a much-needed growth trajectory towards immunity healing to fight infection (6).

7
• Concurrent infection associated with stunting increases household expenditure on healthcare (14). This often causes the affected families' already be facing difficulties.
• If the disease and infection are contagious, other children and adults may be at risk, increasing the household cost for caring for the sick child as well as the parent/s and child's attendants' wage loss (14).
In the retrospective studies (14,24,25) on stunting and the economic outcome, it is evident that stunting plays a crucial role in economic consequences of an individual, household, and communities. Poor nutrition extends the cycle of poverty and malnutrition, through three main groups:

Figure 1. Three main groups that extends the cycle of poverty and malnutrition
According to the World Health Organization, stunted children earn 20% less than average adult salaries than non-stunted children (26). Malnutrition has a huge economic cost, with annual GDP losses estimated to be in the billions of dollars. According to the World Bank, a 1% decrease

Poor Nutrition
Direct losses in productivity and functioning are not to their optimal level as a result of poor physical health and short height in adulthood related to malnutrition and disease Indirect losses due to low cognitive development and less time spent in school.
Increased healthcare cost for infectious and chronic diseases, and other unfavorable outcomes in later life. 8 in adult height due to childhood stunting results in a 1.4 percent decrease in economic productivity (27).

Child Stunting and Economic Outcomes in SAARC Countries
(5) conducted a cohort study in Brazil, Guatemala, India, the Philippines, and South Africa, which followed a group of children from birth to adulthood. The study concluded that children born with smaller heights as adult had smaller stature, lower BMI, less time in school, lower cognitive abilities, smaller wages, and stunted women themselves had wasted infants. The growth restriction in the mother's womb many a time results in adverse effects on the fetus, sometimes in form of death or if saved it may face some life-threatening complications (4). Growth restriction in the mother's womb frequently has a negative impact on the fetus, sometimes resulting in death or, if saved, some life-threatening complications (4). (28) in the study of 54 countries on demographic and health surveys found that there is an association between maternal stunting and wasted infants. There is a consistent association between perinatal deaths and maternal stunting (29).

Educational and economic consequences of child stunting
Restricted blood supply, nutrients, and frequent infections are the common causes of small stature, but they also damage the brain, which results in delays in motor skills and cognitive function permanently (30). Using the same group as taken by (5) in their study, (31)found that children stunted till 24 months have a strong relationship with the reduction in schooling, i.e. by 0.9 years, delay in school enrollment, a 16% higher risk of failing in class, keeping gender, socioeconomic status, and mother's education constant. Evidence also shows that children stunted between the ages of one and three have poor performance in school and fewer achievements (26).
Short stature and lower economic productivity have a link, as shown in a Brazilian study where a 1% increase in height increases wages by 2.4% (32).

9
In developing countries, linkages were found between demand for labor and labor height (32)(33)(34)(35)(36)(37)(38). The strong association between height and wages can also become more authentic when the study of developed countries shows increased earnings for people with attained height (39, 40).

Cognitive impairment and Stunting:
According to one study (30) undernourished children consistently perform worse on attention, fluency, and memorization tests. Severe undernutrition has adverse effects on a child's neurological wellbeing that may result in impaired cognitive development (41). Chronic undernutrition damages the signal transmitting chemical procedure, hence reducing its speed, finally damaging the motor skill abilities and reducing memory consolidation (42).
The cognitive deficit that occurs in early life results in life-long consequences (5). Indian children between 5 and 7 years and from 8 to 10 years were found to be less attentive and had difficulty in memorizing and learning if they were found to be stunted in childhood (30). A study from Zimbabwe by (43) shows that adults who were stunted till the age of 3 years have a deficit in cognitive ability, in comparison with children who is not stunted (44). According to (45) study, one higher grade in school results in a 9% increase in pay.

Increased risk of chronic disease and Stunting
Stunting due to any cause, including malnutrition, undernutrition, undernourishment, a lack of nutrients, infections, sanitation, contaminated food and water, maternal stunting, and/or any other direct or indirect cause, can make a person a higher risk for some type of chronic disease in adulthood.
Dutch famine studies reveal that children born during and immediately after it when they came into their middle age had a higher risk of some chronic diseases as they suffered from impaired glucose metabolism, heart disease, breast cancer, and obesity (46). Similarly, children born in China in the late 1950s or early 1960s during the Great Chinese famine and Biafrans born during and after the Biafra famine in the late 1960s had an increased risk of chronic diseases like diabetes and hypertension as adult (47,48). Although, famine is an extreme example, it supports our study on how undernutrition and undernourishment of a mother and an infant can lead to chronic disease when that child grows up to be an adult. A less intense cohort study was done by (49), and the finding showed that the children who were born thinner and with lewer BMI, at the age of 32 years had impaired glucose tolerance or were diabetic. (5) study also supports the fact that childhood undernutrition results in a high risk of diabetes, hypertension, and cholesterol concentration in adulthood.

Economy and Stunting: A Conjoint Phenomena
In the review of literature (50) the severe acute malnutrition (SAM) burden states that six countries in Asia together have more than 12 million children suffering from SAM. (51) in their study, done between 1820 and 1860 on the US and several European states' industrial sector performance, says that "deleterious effect of industrialization on workers was visible in their physical stature". (52,53) both study shows that higher the education level, the higher the rate of growth of real GDP. A positive and significant association was found between spending made on education and economic growth (54) A study by (55) found that economic growth, education, and technical progress are connected and positively correlated. Increases in earnings have a positive relationship with additional years of education (56,57).
A health deficit at large can result in economic dilapidation, whether of a household or a country. Chronic health condition have a direct resource cost that includes the cost of medicine, medication, health care, and other health services associated with the chronic disease. Loss of employment comes hand in hand with chronic disease, and sometimes the economic conditions of the household get worse in the event of premature death. Sometimes the expenses are borne by the other household member or family member who has to give up working or attending school so that they can look after the sick person at home. According to DALYs (disability-adjusted life years), chronic disease accounts for 54% of all healthy life years lost (58). Various studies suggest that disease burden, in general, is expected to affect economic growth in particular, and chronic illness reduces the supply of workers caused by morbidity, mortality, and early retirement (8,9,59), as well as lowers the person's productivity (60, 61  (71). They estimated a negative relationship between GDP and stunting in the country. According to (72), malnutrition-related mortality and illness result in a direct loss of human capital and productivity for the economy in ACP countries. On a micro level, a 1% reduction in adult height as a result of childhood stunting represents a 1.4 percent reduction in a person's productivity. The economic impact of malnutrition is estimated to be between 2% and 3% of ACP countries' GDP. As a result, better-nourished youngsters are more likely to start school and, as a result, enter the labor market earlier (73).

Figure 2. Cycle between Stunting and Economic Growth
The study on Papua New Guinea (74) shows that although there is economic growth, the stunting rate is stagnant due to the eating practices of the mother during the conceived period and feeding practices after childbirth. Furthermore, sometimes, short stature is common and stunting  13 goes unrecognized by the health worker. PNG is facing the pressure of non-communicable disease due to the low height and increased income used in healthcare that would have been used for nutritious food (74). (75) did the study between psychological factors and stunting in 137 low/middle-income countries, including the SAARC region, and it was found in the study that approximately 7.2 million cases of stunting were attributed to psychological factors; 3.2 million cases of maternal depression showed $14.5 billion in economic costs, followed by the education of 2.9 million with an economic cost of $10 billion and Intimate Partner Violence (IPV) 2.1 million with an economic cost of $8.5 billion. These can have lifetime economic impact.
An association of early child growth restrictions at age 2 and earlier reduces the likelihood of employment in formal sector jobs among young adults 20 years later (15). (16) did the literature study between children's health and the economy, which suggests that children's health is a potentially valuable economic investment and that greater investment results in better educational attainment, leading to productive skilled labor. This sets a favorable demographic change for any country. Further, the study suggests that safeguarding children's health in their childhood is more meaningful than at any other age because failure to do so can lead to lifetime impairment. The study goes on to state that a poor health mechanism has a vicious circle and results in intergenerational poverty transmission. As measured by all scales, children from low socioeconomic status have higher chances of stunting (76).
The results of the study by (77) indicate that wealth inequality is strongly associated with childhood unpropitious growth and stunting and that the poorest 20% of households are three times more vulnerable to suffering from growth retardation than the wealthiest 20%. Results of the study by (18) indicate that the economic burden of early-life growth faltering is substantial and that children in developing countries lose an average of 0.5 years of educational attainment due to

14
early-life growth faltering. That results in a global economic loss of $176.7 billion and, on average, $1400/child loss of lifetime earnings.
The study by (78) shows evidence that although economic growth took place both in Gujarat and Bihar, in Gujarat the undernourished rate of children declined to around 50%, and Bihar showed a very small change in malnourishment. Hence, effective policies are necessary to reduce stunting.
Household wealth of the family matters at an early age for physical growth, and that the condition of poverty and impoverishment can influence growth delays even beyond the 1,000 day window (79). How the labor market gives an advantage to a person's height is explained by (80).
They found that a correlation between anthropometric measurement as height and economic conditions prevails in the labor market and it plays a very effective role. Malnutrition results in a higher number of morbidities and mortality, higher expenditure on healthcare, and less education in children, causing less productive economic outpour and billions of dollars in economic losses (81). A research review by (5) confirms the association between decreased malnutrition and increased economic productivity. In the Copenhagen Consensus (81) it was proved with several pieces of research that malnutrition results in tremendous economic losses, measured in billions of dollars.

Explanation of Data
Dataset of child stunting prevalence, economic growth, urbanization, government stability index, rainfall, and temperature for a sample of eight SAARC countries namely; Afghanistan, Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan, and Sri Lanka from 1984 to 2018 is used 15 under the pretext that the unavailable data will be interpolated, backcasted and forecasted, through the standard procedure.

Advantages of interpolation
Interpolation is likely to be more accurate than extrapolation as it takes the value within the data. It creates a smooth data effect. If the available data is in some boundary it takes the mean and variance of it and creates similar smooth data (82, 83)

Disadvantages of interpolation
Interpolation is although a simple methodology, but it lacks precision. Every data has different characteristics and can range from random to polycyclic processes. It cannot estimate above maximum or below minimum values (82, 83)

Advantages of backcasting
Backcasting in competitive analysis to improve the understanding of the strategic directions (84-86)

Disadvantages of backcasting
The characteristics that the present data is taking into consideration may lack in past or might have more characteristics in the past which the data is missing out (84,85).

Advantages of forecasting
The primary advantage of forecasting is that it provides valuable information that the researcher can use to make decisions about the future. Forecasting looks into past and real-time data to predict the future (84, 87)

Disadvantages of forecasting
Forecasts are never 100% accurate. The characteristic that it is taking might not be there in the future (84, 87) Table 1 gives us descriptive statistics of all our variables used in the research.

Details of the Variables
Child Stunting: The percentage of children under the age of five whose height for age is fewer than two standard deviations of the median height is known as the prevalence of child stunting. Our data is from the WHO 2018.
Gross Domestic Product (GDP) per capita: For GDP per capita, we used data from the World Bank database. By dividing the total population by the gross domestic product, the gross domestic product per capita are being obtained. Data is in constant 2010 in the US$(In purchasing power parity). We have used the log of GDP per capita for our estimation purposes.

Government stability index (GSI): Data from World Bank database on Worldwide
Governance Indicators which has six components namely; voice and accountability, political stability, and absence of violence/terrorism, government effectiveness, regulatory quality, rule of law, and control of corruption has been used. The missing data were interpolated, back-casted, and forecasted individually component by component. Once the data got up to date, an index was created out of six components and was used in our empirical estimations.
Since food security is primarily dealt by the government and its policy's effectiveness.
Therefore, the improved governance reduce stunting by focusing more on providing good, nutritious, and effective distribution of food to its people, hence reducing the chances of stunting (88). Furthermore, the efficient and effective government will ensure other better-living conditions like, improved WASH, education, healthcare, etc. not only for its urban people but also for the rural population as well to slow down the 'urban transition' which puts pressure on cities and towns and standard of living drops there.
Rainfall and Temperature data: Rainfall and temperature data are from the World Bank Database. The dataset is built from monthly observations and converting them into the year. Square and cube of temperature anomaly are also taken in the estimations as it shows the effect of stunting wearing off at a certain point in time and giving a turning point of the relationship, that is, after certain point these natural effects loses its impact. This means that higher temperature (rainfall) increases, the prevalence of stunting but at a consistent level with the increase in temperature (rainfall), stunting increases with decreasing rate. it gives the phase phenomena and gives increase accuracy of the complex temperature (rainfall) system .
Urbanization: Urbanization refers to people living in urban areas as defined by national statistical offices. It is rate as a proportion of the country's population, taken by the World Bank database.

Model Identification
The empirical model is as follows:  (2): This method establishes the causal association between economic growth and stunting, as well as the reverse causality between stunting and economic growth.
The rationale for the identification strategy of rainfall and temperature is that, first; these variables are exogenous and random. Secondly, the earlier literature review suggests that the child's physiology gets affected by the weather shocks and have a strong likelihood to correlate with stunting prevalence (94,95). Third, stunting is a chronic illness that isn't fully determined until the first two years of life, so it isn't a one-time occurrence (89,96).
According to the theory, current weather anomalies are natural physiological shocks that influence future stunting and, as a result, GDP per capita. An exogenous source of variation in per capita GDP growth is necessary to calculate the impact of GDP per capita on stunting. To generate such variation, we use smooth variation in rainfall and temperature anomalies as IVs. A key characteristic of the SAARC region that makes this estimation plausible is that these countries are highly dependent on agriculture (97). Another reason is that higher temperature tends to increase

20
in vector-borne diseases (98). The rainfall shocks have multiple effects too; crop cycle, drought, flooding, disease are the common elements that are associated with excessive and scarce rainfall.
Hansen & Sargan is a test of over-identification. The null hypothesis that the instruments are uncorrelated with the error term is tested in this way. If the p-value is less than 10%, we can conclude that all instruments are not exogenous and thus invalid. We compare the crucial values through Stock and Yogo (99) to test the weak instruments to the Wald F statistic to verify the instrument's strength. This hypothesis is tested by determining whether the maximum relative bias and size distortions are larger than 10%, 15%, or 20%.
A stunting prevalence observation in a given year, according to (92), is the result of one or more years of economic performance prior to that year, and hence, recommend using a model in which regressors are replaced by their five-year moving averages. Eq1 can so be represented as: Where the superscript denotes the five-year rolling average of the regressors; the log of GDP, urbanization, and government stability Index. Each regressor has been calculated taking the current time and time of four previous years. 5 years moving average of GDP is calculated as:

4-RESULTS
The association between economic growth and the prevalence of child stunting is estimated in the study using Stata 13. We conducted stepwise estimations (Table 2.) to reveal the best estimation approach for the research 2 .

Detail of stepwise estimation are available in the supplementary material
Our data comes from eight countries, totaling 280 observations; each country has 35 years of observations. We used a panel data approach for our estimation. Table 3 shows the effect of stunting on economic growth using the ordinary least square method (OLS) taking the dependent, independent, and instrumental variables that we will be using for our two-stage least square estimation (2SLS). p=0.0439 Source: Author's construction and dependent variable is child stunting (Yit) Notes *, **, ***: significant at 10%, 5% and 1%. #: insignificant (p-value >15%), ##: marginally insignificant (p-value<15). Countryclustered t-value in parentheses. Anderson-Robin p-value in Square bracket    (Sit) and for the reverse causal the dependent variable is GDP (lnGDP) Notes *, **, ***: significant at 10%, 5% and 1%. #: insignificant (p-value >15%), ##: marginally insignificant (p-value<15). Country-clustered t-value in parentheses. Anderson-Robin pvalue in Square bracket To determine whether there is reverse causality between our two variables, we employed the 2SLS approach. The post-estimation test for weak, valid, and endogenous instruments is inferable in the 2SLS approach, which allows us to accurately comprehend the result and draw a better inference. The precision of the conclusion is only regarded through the interpolated data that we calculated, which was also a restriction due to the shortage and variability in accessible stunting data. Using country-clustered standard errors, the result is used to test robustness for heteroscedasticity. Table 4 presents estimates for the impact of child stunting on current GDP per capita, demonstrating that GDP per capita is clearly endogenous for child stunting. In Table 4, column 1, the first-stage estimations employ the 2SLS approach to examine the relationship between rainfall and temperature anomalies and child stunting. Rainfall levels that are higher than usual have been associated with a decrease in the prevalence of stunting. According to a coefficient for the beguilement in the temperature shock, the relationship between temperature and stunting is not linear, and small-scale negative temperature shocks increase stunting prevalence and vice versa.

Two-Stage Least Square estimation
An increase in child stunting prevalence reduces GDP per capita by 3.4 percent, according to the estimate in column 2 of Table 4.
OLS estimations of the impact of GDP growth on stunting are expected to be negative due to the negative reverse causal effect. The F-statistics indicate that weak instruments are not a concern in the estimation in Column 2 because they are significantly more than the Stock-Yogo threshold value of 20%. Durbin and Hausman's low probability values are less than 10%, implying that GDP per capita is exogenous. The p-values for the Hansen and Sargan tests in column 2 are greater than 10%, indicating that the tests do not rule out the possibility that the instruments are uncorrelated with the second-stage error. We re-estimated Equation 2 in columns 3, 4, and 5 by including rainfall and temperature anomalies. The coefficients demonstrate that the link between stunting and economic growth is nonlinear, i.e., stunting reduces as economic growth increases. Source: Author's construction and dependent variable is GDP (ln GDP) Notes *, **, ***: significant at 10%, 5% and 1%. #: insignificant (p-value >15%), ##: marginally insignificant (p-value<15). Countryclustered t-value in parentheses. Anderson-Robin p-value in Square bracket Table 5 shows estimates of the influence of GDP growth on child stunting. The 2SLS

GDP on child stunting effects
coefficients are greater than their OLS equivalents, indicating that negative reverse causality exists. All of the 2SLS coefficients are negative, significant at 1% in column 2, and imply that a 10% rise in GDP per capita would result in a 6% 3 reduction in stunting prevalence.
Lastly, we found that urbanization has a positive and mostly significant impact of 1% on stunting, which implies that with more urban transition due to the factors mentioned in the literature review, it is evident that urbanization plays quite a part in stunting prevalence.
Governance, although it mostly has a negative coefficient that implies that good governance leads to a decrease in stunting, is insignificant with the high p-value. Rainfall shocks do not affect stunting in any of our estimations, while temperature shocks have a negative impact on stunting.

Alternative model using 5-Year moving averages
We have reproduced Equation 5, in which the regressors are given as 5-year moving averages. This estimation has the advantage of assisting us in determining the impact of, for example, this year's stunting as a result of the previous four and current year's GDP. The problem with this estimation is that we may not be able to compare this with Table 3 because Equation 5 implies a different structural model. Nonetheless, if the result shows similar evidence in estimations, it would give reassurance to our estimations of previous analysis but still we cannot ignore the fact that stunting data is been interpolated due to scarcity and inconsistency in available data.
Our estimations in Table 6 on 5-years moving average also indicates that stunting affects economic growth and is highly significant. In the same line Table 7 shows that economic growth also plays an important role in stunting and is highly significant in 5-years. Country fixed effect Yes n.a n.a n.a n.a n.a Year Fixed effect Yes n.a n.a n.a n.a n.a R² 75

5-DISCUSSION
Human capital is priceless and is measured by health and education in a country. Human capital that is healthier and more efficient contributes to greater economic growth. Similarly, less skilled and unhealthier human capital is more likely to become a burden. This inefficient productivity problem has an impact on a country's GDP. Stunting in childhood has an impact on a person's health and productivity in adulthood. Expenses for weakened immunity, chronic sickness, health transmission to the next generation, and other factors may exacerbate a person's financial condition.
Despite the fact that the SAARC region's human capital is among the top ten out of 195 countries, there are significant differences in human productivity levels. Stunting costs low-income developing countries, especially those in South Asia, up to 3% of their GDP. Stunting results in the biggest human capital losses in this region and can have irreversible, long-term consequences that impede economic growth, resulting in a significant impact on the country's GDP. According to the estimates, stunting lowers per capita income. Stunting in childhood has been linked to lower school attendance. This could be due to inadequate motor and cognitive skills, frequent infections and illnesses, chronic disease, and other stunting-related challenges. Adults with little education and health issues have a more difficult time obtaining work in a competitive economy. If they do find jobs, their performance is lower than that of a person who was not stunted in childhood due to their inferior physical and mental capacity.
Faster economic growth, as indicated by developed countries, lessens stunting. Economists relate this to rising per capita GDP. Macroeconomic shocks have an impact on income, which in turn has an impact on generations. Institutions attempting to minimize stunting prevalence must improve policies and interventions such that GDP or national income growth has no or little influence. Furthermore, policies must reach a bigger number of effected individuals.
The research presented the effects of rising current GDP per capita on child stunting in the SAARC area, which comprises Afghanistan, Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan, and Sri Lanka. The 2SLS is a method for determining the reverse causal link between dependent and independent variables, and we were able to do so successfully. Our data suggests that a 10% rise in GDP per capita reduces the frequency of child stunting by 6%. (6.7 percent by the five-year moving average method).
The comprehensive analysis also reveals that our estimate is in the same ballpark as our previous estimates. We also calculate the GDP per capita effects of stunting in the reverse causal direction. And the results show that every percentage point increase in child stunting causes a 3.5 percent drop in current GDP per capita.

6-CONCLUSION
Stunting has reasons and consequences attached to it. It affects the child's mental and physical health, which has an impact on its productivity in every aspect of life. This productivity lag affects its income earning in adulthood. If seen from another angle, the expenditure made on the stunted child has an impact on the other siblings' food, health, and education along with the earner's income, which drifts the household into a deeper poverty level. The literature suggests that there is an inverse relation between poverty and GDP per capita. Hence, the more nonproductive and poor the nation, the lower the GDP of that country.
The WHO's sustainable development goal (SDG) includes a major focus on how to reduce child stunting. Stunting and all forms of undernutrition are the objectives that they want to curb by 2030 by all means. Research has proven that economic growth can be a key to achieving the goal.
While studies prove that economic growth can play a role in the development of every country, This pretext gave the researchers reason to consider how far it can be beneficial for lowering stunting prevalence, as economic growth is essential in every way. The empirical results gave mixed results as far as the magnitude of the impact is concerned. The magnitude of the result can vary as the data is interpolated due to the scarcity and inconsistency of data, and this is the limitation that we faced.
We found that economic growth is moderately pro-poor. Also, the study suggests that economic growth is not the only means that is enough for the reduction in stunting prevalence; other supportive attributes are also needed. If any SAARC country is serious about this curse, they need to focus on directional, focused, effective, and efficient interventions. The estimation was unable to find any significant relation between governance and the decrease in stunting.
Our research can be of great help to policymakers in giving directions to the strategies that they use to influence stunting in the downward direction. Policymakers should make sure that the strategies they suggest are pro-poor.
Although this approach can be beneficial in other development sectors, if the right instrument is identified, the result will be unbiassed. We can extend this research further by using other instruments, such as trade openness, education, etc., to see how they affect our results. We recommend that more research in line with this can give a clearer picture of how the reduction in stunting is workable in better ways. This approach also has its limitations, as it cannot be applied to smaller areas or places where the data is completely unavailable. An extensive amount of data