GDP, health expenditure, industrialization, education and environmental sustainability impact on child mortality: Evidence from G-7 countries

ABSTRACT The child mortality rate for children under five years of age is a major problem worldwide. The purpose of this study is to investigate the factors responsible for controlling the under-5 mortality rate in G-7 countries. The study utilized monthly time series data from 1971 to 2021 and employed Grossman’s health production function with Driscoll-Kraay’s standard error estimation approach for all statistical analyses. The results showed that economic expansion and the use of renewable energy contribute to the reduction of under-5 mortality rates in G-7 countries, whereas fossil fuels, industrialization, education, and unemployment contribute to the increase in mortality rates. The robust results also revealed that GDP, renewable energy, and education help mitigate under-5 mortality rates in G-7 countries. This analysis plays an essential role in helping G-7 countries reduce their under-5 mortality rates while considering environmental sustainability. Additionally, the required discussions and policy recommendations are also presented.


Introduction
The number of infants and children under the age of five who die per 1,000 live births is referred to as the underfive mortality rate (U5MR).The under-five mortality rate (U5MR) is an important indicator of a society's health.The predictor of under-five mortality is critical for developing successful prevention programs that improve young children's survival.Mortality rates are regarded as a crucial indicator in the interpretation of a country's population demographics; thus, the United Nations Children's Emergency Fund (UNICEF), the World Health Organization (WHO), the World Bank, and the United Nations Population Division (UNDP) allied in 2004 to further the work on monitoring the behavior of child mortality series over time (You et al., 2015).Even in developed countries, U5MR is favored over IMR because the U5MR series enables a broader age range, from the neonate (a few days after birth), infant (one month after birth to 12 months), and child (from 12 months after birth to 59 months after birth).Throughout that time, a child is exposed to a diverse range of illnesses and ailments that can be fatal.In general, child mortality rates affect life expectancy.This study selected G7 countries for empirical analysis because, within the last 50 years, the G7 countries' life expectancy has grown (Acsadi & Nemeskeri, 1974;Babel et al., 2008;Haleem et al., 2008;You et al., 2015).France, the United States, the United Kingdom, Germany, Japan, Italy, and Canada are countries included in the G7 region.Under-5 mortality in Canada was 4.98 deaths per 1,000 live births in 2020.The Canadian under-5 mortality rate decreased dramatically from 20.9 deaths out of every 1000 live babies in 1971 to 4.98 deaths for every 1000 live births in 2020 (Knoema, 2023).The IMR in France, which was already at a low level in 2003, has shown a further decline in subsequent years.In that particular year, a mortality rate of four per thousand live births was observed.The rate had declined over the course of several years, ultimately hitting the lowest point in 2011, 2012, and 2014, with a value of 3.3.In the year 2023, there was a significant rise in the rate, which reached a value of 3.7 (Statista, 2023b).The prevailing IMR in Japan stands at 1.6409 newborn deaths per 1,000 live births.According to projections, Japan's IMR is expected to decline to 0.3328 newborn deaths per 1,000 live births by the year 2100 (Database.earth, 2023).Furthermore, it is noteworthy to mention that the IMR in the United Kingdom had a decline from 4.3 in 2011 to a significantly lower rate of 3.7 in 2021 (Statista, 2023d).Moreover, IMR in Germany and Italy also declined significantly in 2022 (Statista, 2023a(Statista, , 2023c)).In the year 2020, the United States exhibited an IMR of 5.4 fatalities per 1,000 live births, surpassing the rates observed in other industrialized nations (The Commonwealth Fund, 2023).During the period spanning from January 2020 to December 2021, a decline in life expectancy was observed in the United States as well as several other nations (Kuehn, 2022).
The area of infant mortality rate encompasses various factors that have substantial influence, with the emergence of seven prominent features that hold unique significance in distinct nations and households.The indicators being considered include gross domestic product (GDP), health expenditure, renewable energy, fossil fuel consumption, unemployment rate, industrialization, and education.So, the primary focus of this paper is to figure out how the selected variables helped the G7 countries to attain a low rate of infant mortality.So, the research question is whether the impact of GDP, health expenditure, renewable energy, fossil fuel consumption, unemployment rate, industrialization, and education on infant mortality rate is positive or negative.According to Safi and Fatima (Shafi & Fatima, 2019), rising life expectancy was accompanied by rising GDP per capita in G-7 countries.Healthcare spending is an indispensable part of alleviating the mortality rate under 5 in any country.According to Owusu et al (Owusu et al., 2021), a 1% boost in health expenditure lowers infant mortality by 0.19 to 1.45%.In comparison to countries with high infant mortality rates, nations that have low infant mortality rates are better able to respond to healthcare spending.
According to the WHO Report (Joint Monitoring Program JMP, World Health Organization WHO, & United Nations Children's Fund UNICEF, 2020), biomass fuels are used by more than half of the global population (52%), as well as more than 2.4 billion people in third-world countries who depend solely on biomass consumption.They cook with open-fire cookstoves and use flame lamps to light their homes inefficiently (Coffey et al., 2021).The combustion of biomass in an open-fire cook stove generates smoke containing potentially damaging particles for human health (Maher et al., 2021).Renewable energy enhances human health (Caruso et al., 2020) by promoting environmental restoration, reducing water pollution, and increasing biodiversity (Haines et al., 2007;Hanif, 2018;Wang et al., 2019).The effects of industrialization on infant health can be both negative and beneficial.Manufacturers release pollutants into the atmosphere and water; these carbon outputs are frequently low in developed countries.Air pollution is a significant component of the most common cause of infant mortality, acute respiratory infections (Romieu et al., 2002).Working in the opposite direction, industrialization has the potential to improve residential satisfaction, hygiene, and domestic pollution, all of which can improve health outcomes.Likewise, rising income levels can promote nutrition and the availability of health care for mothers and children, lowering death rates (Fogel, 1994).
Focusing on G-7 countries, the average unemployment rate for 2021 was 5.96 percent.Italy had the highest value at 9.83 percent, while Japan had the lowest at 2.8 percent (TheGlobalEconomy.com, 2023).Children born while their parents are unemployed are lighter (De Cao et al., 2022;Högberg et al., 2021;Lindo, 2011).Lower birth weight is associated with lifelong detriments in both health and economic advantages (Behrman & Rosenzweig, 2004;Black et al., 2007;Persson & Rossin-Slater, 2018;Torche, 2018;UNICEF, & WHO, 2019).The impact of job loss on childbearing outcomes is most likely due to maternal stress, rising alcohol consumption, and a lack of adequate nutrition and postnatal healthcare (Aizer & Currie, 2014;Lindo, 2011).So, the unemployment rate has a detrimental impact on child mortality under age 5. Education is suggested to improve a mother's understanding of medical techniques such as contraception use, nutritious food, infection control, and preventative medicine (Mosley & Chen, 1984b).Education is also a crucial factor in socioeconomic development, and it is strongly associated with a variety of socioeconomic status measures, including personality, income, and reproduction for both individuals and societies (Mosley & Chen, 1984b).Maternal education is strongly related to infant mortality (Asefa et al., 2000;Cochrane et al., 1980;Shamebo et al., 1993).Maternal education, thus, according to social science researchers, is considered one of the most crucial factors of infant mortality (Abuqamar et al., 2011;Desai & Alva, 1998;Mosley & Chen, 1984a, 1984b).Thus, the ongoing study investigates the effects of GDP, health expenditure, fossil fuel energy, renewable energy consumption, industrialization, unemployment, and education on the under-5 mortality rate in G-7 countries.
This study specifically examines the time series of death rates only for the G7 countries, as these nations are characterized by advanced levels of development and substantial population sizes.There are variations in the Under-5 Mortality Rates (U5MRs) among these nations.The variations observed among nations are contingent upon the level of development within each country.The study's purpose is to examine the effect of GDP, health expenditure, renewable energy, fossil fuel, unemployment, industrialization, and education variables on infant mortality and determine whether the relationship is positive or negative, as well as what policies should be implemented to solve child mortality challenges.The significance of this variable as a measure of a child's welfare.This is the earlier study to examine the influence of these variables on infant mortality rates in G7 countries.Tests for cross-sectional dependence, unit root tests, and cointegration tests were a part of this analysis to assess the prevalence of static data.After that, the model with the Driscoll-Kraay standard error approach is utilized for the long-run estimation findings, and the AMG model is employed to test the robustness of the estimations.

Literature review
Sial et al (Sial et al., 2022) studied the relationship between fossil fuel consumption and infant mortality rates in 15 Asian economies from 1996 to 2019.They discovered that the use of fossil fuels is lowering living standards.The study concludes that fossil fuel energy consumption has a U-shaped relationship that explains infant mortality.The findings indicate that excessive use of fossil fuel energy is lowering living standards in Asian economies due to low air quality levels.Asif et al (Asif et al., 2022).used multivariable logistic regression to examine the socioeconomic factors that contribute to child mortality and found that child mortality decreased with an improvement in women's education, empowerment, their husband's education, the wealth status of their households, access to clean drinking water, access to toilet facilities, and exposure to mass media.Kiross et al (Kiross et al., 2019) investigated the impact of maternal education on infant mortality in Ethiopia.They discovered that maternal education has a longterm impact on alleviating infant mortality.Cardona et al (Cardona et al., 2022) investigated the consequences of the economic recession on the mortality rate of children under the age of five in 129 countries.They discovered that economic downturns in low-and middle-income countries (LMICs) can increase child mortality by influencing nutritional, physical, and careseeking factors and that recessions resulted in higher losses of under-5 lives.
Adeleye et al (Adeleye et al., 2023).looked into the effects of carbon emissions and non-renewable energy use on infant and under-5 mortality rates in Europe.Carbon emissions worsen infant and under-5 mortality rates, while non-renewable energy had mortalitylowering properties, according to system GMM and quantile regressions.Majeed et al (Majeed et al., 2021) examined Renewable Energy Consumption and Health Outcomes by applying the SGMM, 2SLS, and fixed effects.They discovered using renewable energy raises life expectancy and lowers mortality rates.The favorable connection between clean energy and human health implies that clean energy aids in the control of chronic diseases, resulting in a high life expectancy and low mortality.Wehby et al (Wehby et al., 2017).examined how unemployment cycles impacted child and maternal health in Argentina from 1994 to 2006.They discovered that Argentina's rising unemployment has had a detrimental effect on certain infant and maternal health outcomes.Jakovljevic et al (Jakovljevic et al., 2020) used a two-way fixed effects model to examine healthcare expenditures in the leading Asian countries from 1996 to 2017.The findings concluded that in non-OECD Asian countries, population and urbanization are significantly linked with a longer life expectancy.In both sub-samples, there exists a significant negative association between GDP per capita and infant mortality.Unemployment is inversely related to infant mortality in Japan and South Korea.Shin & Sangsoo (Shin, 2019) investigated how the unemployment rate influenced infant mortality rates in the United States from 2000 to 2016.According to the SGMM findings, every 1% increment in the unemployment rate was associated with a 0.019 rise in infant mortality.Pieters & Rawlings (Pieters & Rawlings, 2020) examined the impact of both maternal and paternal unemployment on children 's health in China using panel data from 1997 to 2004.The findings of the fixed effect model revealed that paternal unemployment has a bad impact on child health, whereas maternal unemployment has a favorable impact on child health.According to Bhalotra (Bhalotra, 2006), Erdoan et al (Erdoğan et al., 2013), and Morgado (Morgado, 2014) countries' mortality rates decrease as their real per capita GDP rises.Koengkan et al (Koengkan et al., 2021).used a panel quantile regression approach on data from 1990 to 2016 to demonstrate that renewable energy had both direct and indirect effects on mortality rates.According to Demetriou and Tzitziris (Demetriou & Tzitziris, 2017) GDP per capita and infant mortality rate increased with increasing returns to scale.That is, at the highest level of income, infant mortality falls until it reaches its lowest point, at which point it begins to rise.
Tejada et al (Tejada et al., 2019) investigated the effect of the economic downturn, unemployment rate, and health expenditure on infant mortality from 1995 to 2014.Lower GDP, unemployment rates, and misery index were discovered to be related to significantly greater child mortality rates in the model with fixed effects.A relatively high proportion of public health spending, in contrast, mitigates the impact of financial indicators on child deaths.Houweling et al (Houweling et al., 2005) conducted a cross-sectional study in 43 developing countries to investigate the factors that influence the under-5 mortality rate.They discovered that higher national income was linked to minimizing under-5 mortality rates.Steadily increasing public spending on health could help to mitigate this effect.Wang et al (Wang et al., 2016) discovered that advancements in education attainment and economic expansion had made a significant contribution to a decrease in child mortality; more than 60% of Chinese counties had rates of a declining trend in under-5 mortality rates that were significantly quicker than expected.Using the ARDL method, Asumadu-Sarkodie et al (Asumadu-Sarkodie et al., 2016) investigated how the fertility rate, GDP, household final consumption expenditure, and food production affected child mortality in Ghana.They discovered that rising levels of social determinants such as GDP and household final consumption expenditure contributed to lower child mortality rates in Ghana.To reduce child mortality rates among children under the age of five, infants, and the vulnerable in Ghana, hunger must be eliminated and access to safe and nutritious food must be ensured.However, the literature gap indicated that analyzing the existing literature, it is observed that in G-7 countries there are no studies that analyze the impact of GDP, renewable energy, industrialization, and education on infant mortality.Moreover, some researchers except for G-7 countries analyzed the impact of the mentioned variables but they didn't apply the most famous Driscoll-Kraay (Driscoll & Kraay, 1998) approach.So, this paper is crucial for the context of the G-7 countries to fulfill the gap using the Driscoll-Kraay method.

Theoretical framework and health production function
To appropriately articulate the purpose of this paper is to investigate the impact of GDP, health expenditure, fossil fuel, renewable energy, industrialization, unemployment, and education on the under-5 mortality rate.Analyzing the nature of the dependent and independent variables of this paper, the model that is perfectly matched is called the Health Output (HO) theory.The HO model was proposed by Grossman (Grossman, 1972).The health production function, which Grossman (Grossman, 1972) pro-poses, explains the connection between health input and an individual's health output.The concept that health is determined by a range of circumstances, some of which may be induced by the individual, meaning that health can be generated, is the foundation of the theoretical health production function put forth by Grossman (Grossman, 1972).The following Equation 1 is an explanation of the production of individual health: Where, HO stands for an individual's health output and HI for a person's health-related input.The aforementioned methodology examines the micro level of individual health outcomes.This study translated the aforementioned model to the macro level by Majeed and Ozturk (Majeed & Ozturk, 2020) study.The following Equation 2 can be used to express the economic, environmental, and social elements (Majeed & Ozturk, 2020;Siddique & Kiani, 2020) that affect health: This study has included a variety of economic, healthcare, and social elements in our models in addition to environmental considerations: GDP, health expenditure, fossil fuel, renewable energy, industrialization, unemployment, and education.Thus, the following Equation 3 is the model for this empirical research: Where, MR is our dependent variable that represents the mortality rate under 5.  (Tejada et al., 2019) among others.The coefficient estimates will provide the elasticities directly is one of the key benefits of using the logarithm.Consequently, the following Equation 4 will be the logarithmic transformation for this empirical study: Moreover, recently this HO model is used by Azam et al (Azam et al., 2023) to investigate the relationship between life expectancy and environmental degradation.Similarly, the other researchers such as Rahman et al (Rahman et al., 2021), Radmehr & Adebayo (Radmehr & Adebayo, 2022), Liu et al (Liu et al., 2022), Tariq & Xu (Tariq & Xu, 2022), Liu (Liu, 2022), and Xu et al (Xu et al., 2022).also used the HO theory in their analysis to specifically relate the dependent and independent variables.

Data and descriptive statistics
Table 1 provides a comprehensive overview of the variables under investigation in this study, along with their respective abbreviations, descriptions, measurements, and sources of data.The primary focus is on figuring out the complex factors that contribute to infant mortality rates throughout the G7.shows that all of the variables are stable, no unstable variables exist.It also narrated that the value of renewable energy is the lowest, while the value of industrialization is the highest.

Cross-sectional dependence (CSD)
The research examines the panel data for crosssectional dependence, serial correlation, and heteroscedasticity.To determine whether there is heteroscedasticity in the data set, modified Wald statistics for groupwise heteroskedasticity will be utilized (Tufail et al., 2021).Using De Hoyos and Sarafidis (De Hoyos & Sarafidis, 2006) the existence of serial correlation will be examined.Pesaran (Pesaran, 2004) developed a diagnostic method for determining if cross-sectional dependence exists is the CSD statistic.
Pesaran (Pesaran, 2004) CSD test is given as below in Equation 5: ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi Where ρ2 ij represents the pairwise cross-sectional correlation coefficient of residuals, and T and N represent the time and cross-sectional dimensions of the panel, respectively.In this setting, the null hypothesis has cross-sectional independence with CSD ~ N (0, 1).
The cross-sectional dependence tests of null hypothesis states that each unit is independent of the others.Traditional unit root tests, such as those developed by Kao and Chiang (Kao et al., 1999) and Pedroni (Pedroni, 2001), may yield incorrect results due to slope variability and cross-sectional dependency (Dogan et al., 2017).As a result, this paper used Pesaran's (Pesaran, 2007) CIPS, a second-generation unit root test, to determine whether the variables in the CSD existence and slope heterogeneity were stationary.A cross-sectional average of ti is required to calculate a CIPS estimate, as shown in the accompanying illustration in Equation 6: CIPS has gained popularity in recent studies because of its capability to resolve CSD and heterogeneity.This test's null hypothesis is that the series in question appears to contain a unit root.As a result, if the variable is stationary at the first difference, a cointegration test should be performed prior to parameter estimation.

Panel unit root test
First-generation panel unit root tests, such as Levin-Lin Chu (LLC), ImPesaran-Shin (IPS), augmented Dickey-Fuller (ADF), and PhillipsPerron (PP), are invalid due to cross-sectional dependence (Pesaran, 2007).As a result, Pesaran (Pesaran, 2007) developed second-generation panel unit tests that are reliable in the presence of crosssectional dependence, such as the cross-sectionally augmented Dickey-Fuller (CADF) and the cross-sectionally augmented Im-Pesaran-Shin (CIPS).The CADF statistic can be conducted in the following Equation 7: Cross-sectional averages of lagged levels and first differences of individual series are shown in Y tÀ 1 and ΔY it , respectively.By averaging the CADFi, the CADF statistic can be calculated as follows in Equation 8: CADFi denotes the t-statistics in the CADF regression defined by Equation 8.

Cointegration testing
Because of presence of CD, 1st generation co-integration test developed and used earlier (Larsson et al., 2001;Pedroni, 2004;Westerlund, 2005) completely fail to predict size distortion properties in the panel data.Even Kao et al. (Kao & Chiang, 1998) and Pedroni (Pedroni, 2001) did not account CD amid cross-section under considerations.Therefore, heterogeneous estimation approach is used to find the presence of co-integration because of existence of CD, heterogeneity, and non-stationarity problems in the data.Westerlund and Edgerton (Westerlund & Edgerton, 2008) method incorporates heterogeneity in the slope, CD, and correlated errors.This study used the second-generation (Westerlund & Edgerton, 2008) panel cointegration approach to determine the cointegration connections between the variables of concern.This technique accurately predicts the cointegration properties in cross-sectionally dependent heterogeneous panel data sets.It also calculates four-panel non-cointegration test statistics based on error correction.This test is generally defined as follows in Equations 9, 10, 11, and 12: The G t and G α stand for group means statistics in equation 9 and 10, while Y t and Y α stand for cointegration in equation 11 and 12.These test statistics are anticipated when the null hypothesis is that there is no co-integrating link between the variables in a model and the alternative hypothesis is that there are co-integrating relationships.

Driscoll kraay standard error approach (DK)
In this study, the Driscoll & Kraay (Driscoll & Kraay, 1998) standard error approach is employed to identify the long-run effect of health expenditure, fossil fuel, renewable energy, GDP, industrialization, unemployment and education on under-5 mortality rate.This strategy is seen to be the best if the data exhibit heteroscedasticity, CSD because it is non-parametric and provides for flexibility and a large run dimension.And this is the main model of this study and the required result discussion and policy formulations will be based on this method.Panel data typically exhibit heteroscedasticity, serial correlations, and cross-sectional dependences due to factors such as increasing data accessibility, rapid urbanization and industrialization, prioritization of improving the research and production sectors, better educational opportunities, positive economic growth, significant industrial pollution, attention to public health issues, and economic globalization.In the case of the most developed and developing nations in the world, all these variables are more prevalent.Additionally, if the cross-section dimension (N) and the time series dimension (T) are both large, the economic development of the various nations may be interdependent.Because of this, disregarding heteroscedasticity, serial correlations, and cross-sectional dependences may result in ineffective statistical inference (Qiu et al., 2019).When heteroscedasticity, autocorrelation, and cross-sectional dependency are present in a panel data set, the usual fixed effect model will not be able to generate reliable results.As a result, the STATA xtscc program's Hoechle (Hoechle, 2007) process, which produces Driscoll & Kraay (Driscoll & Kraay, 1998) standard error technique for linear panel models, is used in this investigation.These are consistent for heteroskedasticity and also robust to general forms of cross-sectional dependence to examine the impact of pollution on health status for a panel of developed, developing and poor countries.The advantages of DK standard error are considered one of the best techniques if there is any chance of heteroscedasticity and spatial and serial dependency in the data (Özokcu & Özdemir, 2017;Sarkodie & Strezov, 2019).Thus, this study utilizes DK standard errors for pooled ordinary least squares (OLS) estimation by considering a linear model expressed in Equation 13: Where, y i; t is the dependent variable (mortality rate) and x i; t denotes the explanatory variables (Current health expenditure, fossil fuel energy, renewable energy, GDP, industrialization, unemployment, and education).

AMG, MG, CCEMG
To assess the robustness of the Driscoll-Kraay approach, this study additionally utilized the Augmented Mean Group (AMG).The AMG methodology has significance in establishing the robustness of the Driscoll-Kraay approach and validating the reliability of its conclusions.Traditional panel regression methods may be biased and inconsistent when cross-sectional dependence is present (Paramati et al., 2017;Pesaran & Smith, 1995;Phillips & Sul, 2003;Sarafidis & Robertson, 2009).
AMG estimator introduced by Eberhardt and Bond (Eberhardt & Bond, 2009) and Eberhardt and Teal (Eberhardt & Teal, 2010), in addition to CCEMG, is highly robust in the face of cross-sectional dependence and slope heterogeneity.The unobservable common factors ft specified in Equation ( 10) by the common dynamic effect parameter are captured by the AMG estimator.Consider the first difference OLS equation to describe the AMG estimator shows in Equation 14: The AMG estimator is then calculated using the groupspecific parameters is shown in Equation 15that have been averaged across panels: The formula for the AMG estimator is given in Equation 15.It is derived from the estimates of βi that are the estimates of β i in equation 14's description of an OLS regression at first difference, where Δ and θ are the difference operator and the coefficients of the time dummy D, respectively.Thus, the AMG estimator is used to check for robustness.

Results and findings
The first step in panel data analysis proposed by Pesaran (Pesaran, 2007) is to check for cross-sectional dependency in the data.This step is critical because it gets to decide whether first-generation or second-generation models can be employed in data processing.The Pesaran CD, Frees Q statistics, and Friedman tests' results reject null for the models (Table 3).There is cross-sectional dependence in mortality rate, health expenditure, GDP, industrialization, education, fossil fuel energy, renewable energy, and unemployment.Table 4 depicts the findings of these unit root tests, which suggest that even some variables are stationary at the first difference, or I (0), while some others were becoming stationary after only one difference, or I (1).To summarize, variables employed in interpretation are either I (0) or I (1), however, none of these are I (2).
Table 5 shows the results of the panel cointegration test accomplished by Westerlund and Edgerton (Westerlund & Edgerton, 2008).The conclusions are based on both null and alternative hypotheses, with the aforementioned noting cointegration among the variables showed by group means statistics and also consider the establishment of CSD and structural breaks.The findings indicate that this paper rejected the null hypothesis of no integration among study variables.It would support the contention that the interest variables have a cointegration relationship.
The long run estimation result of this paper in G-7 regions is demonstrated in Table 6.This table shows that if economy booms by 1% then child mortality reduces by 0.044%.That is, economic growth is really beneficial for G-7 countries to reduce their child death.Houweling et al. (Houweling et al., 2005), Wang et al. (Wang et al., 2016), Asumadu-Sarkodie et al. (Asumadu-Sarkodie et al., 2016), Jakovljevic et al. (Jakovljevic et al., 2020) narrated that higher economic growth reduces under 5 mortality rates.But heath expenditure reflects a positive sign, suggesting that a health expenditure has been failed to reduce child mortality rate.Owusu et al (Owusu et al., 2021) contradicted this outcome and demonstrated that health expenditure lessens the child mortality rate.By contrast, renewable energy helps to reduce child health vulnerability by reducing child mortality rate that means a 1% increase the usages renewable energy suggests a 0.1240 % reduction of the child death rate.Majeed et al (Majeed et al., 2021) explained that renewable energy reduced the mortality rate.By contrast, fossil fuels also increase the infant mortality rate by reducing the standard of living, creates environmental vulnerability, and health hazard.That shows that if fossil fuel consumption rises by 1%, then child mortality rate rises by 0.2199%.Sial   et al (Sial et al., 2022).also narrated that fossil fuel enhanced the infant mortality rate.Industry makes the environment severe and raises the child's death percentage rate, narrating that if industrialization rises by 1%, it causes a 0.0189% rise the child's mortality rate.In addition, unemployment also rises the child mortality rate, as it shows a positive relation with child mortality rate.A 1% rise the unemployment rate illustrates a 0.1318% rise the child's death rate.Wehby et al (Wehby et al., 2017) in Argentina, Jakovljevic et al (Jakovljevic et al., 2020) in Asian economies, Shin & Sangsoo (Shin, 2019) in USA, Tejada et al (Tejada et al., 2019) illustrated that unemployment badly impacted the infant mortality rate.Education also increases the child's death percentage rate, suggesting that, if education increases by 1% then child mortality rises by 0.8903 %.This outcome is contradicted to Wang et al (Wang et al., 2016).& Asif et al (Asif et al., 2022), where, they concluded that education alleviates the infant's mortality rate.Observing the t values, it is seen that only renewable energy, fossil fuel and education variables are statistically significant, whether all other variables are statistically insignificant.
Table 7 shows the robustness result for the AMG model.The robustness result of the AMG reflects that GDP has a negative relation with child mortality, and the coefficient is −0.850.An increase of 1% in GDP can help reduce the child mortality rate by 0.850%, but this result is not statistically significant.The result of AMG also shows that renewable energy suggests a negative relation with child death rates, and the coefficient is −0.0201, which is also statistically significant.The result of AMG illustrated that health expenditure failed to alleviate the mortality rate under 5 because of lower health expenditure.The result of the AMG of fossil fuel also shows a statistically significant result and a positive relation with the child mortality rate, and the coefficients are 0.556.The education also shows a negative relation, and the coefficients are −0.0476;so, it is seen that the results are robust.More education can reduce the child mortality rate; a higher GDP and the use of renewable energy also help reduce the rate.

Discussion
Table 6 shows that GDP reduces the infant mortality rate under 5.Because rising GDP typically means that people spend more, that more jobs are created, that more taxes are paid, and that workers receive higher pay raises, When people's incomes rise, they can properly take care of their children by providing better health care, nutritious food, and a suitable environment, which consequently reduces the infant mortality rate.Health spending failed to reduce infant mortality because health services in the G-7 countries health-care costs are high and discouraging people from taking medications.Renewable energy keeps the environment carbonfree, and the potentiality of the renewable sources is endless.As renewable energy produces no carbon or less carbon, the carbon-free environment ensures a longer life and consequently reduces the less than 5 mortality rate.Toxin-laden fine particles produced by the combustion of fossil fuels are small enough to penetrate deep into the lungs, causing coronary heart disease, strokes, and premature death.Using fossil fuels makes the environment more deteriorated, and it impacts the infant's child severely.Consequently, using fossil fuels raises the infant mortality rate.Industrialization increases the infant mortality rate, because industrialization means more pollution.The pollution commonly associated with manufacturing has a negative impact on children's health.By creating a harm-weighted index of predicted manufacturing emissions and estimating its relationship with infant mortality, researchers can determine which industries are particularly harmful to infant health.Unemployment also raises the child mortality rate.A high unemployment rate has numerous effects on the economy.Unemployed individuals typically spend less and accumulate more debt, which may result in higher payments.Unemployment causes financial hardship for workers, affecting their families, relationships, and communities (Nahrin et al., 2023;Rahman et al., 2022;Voumik et al., 2023).When this occurs, consumer spending, which is one of an economy's leading sectors, falls, potentially leading to a recession or even depression if not addressed.As a result, infant mortality rises.Education raises the infant mortality rate under 5.Given that the G7 countries consist of industrialized nations, it may be inferred that a significant portion of their respective populations has a high level of education.There is an increasing trend among women to prioritize their professional pursuits and pursue advanced academic qualifications, resulting in a desire to delay childbirth until a later stage in life.Therefore, as a consequence of declining fertility rates, there is an increase in child mortality.

Conclusion
The fundamental purpose of this paper is to investigate the impact of current health expenditure, industrialization, unemployment, GDP, fossil fuel consumption, renewable energy consumption, and education on the mortality rate under 5 in G-7 countries.The most specific objective of this paper was to estimate the long-run dynamics of the selected factors and also to test the H 0 theory.For this analysis, the data were collected from the world development indicators.To stand for the strong theoretical background behind this analysis, Grossman's (Grossman, 1972)

Policy recommendation
Based on the above findings and conclusions in Table 6, the following policy recommendations are suggested for the G-7 countries: • As GDP enhances the under-five child's survival.So, the governments of the G-7 countries should focus on economic prosperity because more economic expansion lowers the infant mortality rate.
Besides, the government along with the private organizations should also work together and highlight the sectors that boom the economy significantly and raise the nation's household standard of living so that the G-7 countries can free themselves from the problems of the under-5 mortality rates.
• The prioritization of enhancing the primary healthcare system carries significant implications for promoting the progress of preventive care, quick identification, and prompt intervention.It is recommended that policymakers dedicate resources towards the development and enhancement of primary healthcare infrastructure, strengthen the capacity of primary care professionals, and prioritize projects that focus on health promotion and illness prevention.This approach ensures that individuals will receive more comprehensive and cost-effective healthcare services, leading to improved child well-being and increased longevity.
• Renewable energy reduces the infant mortality rate.
Renewable energy produces more energy and reduces the scarcity of that energy; as a result, a sustainable environment improves the health condition of the environment, and the under-five mortality rate declines significantly.So, the governments of the G-7 countries should encourage the people to make more use of renewable energy.Also, the government can increase the renewable energy sector by investing more in it.Similarly, the government can also do another task to decrease the cost of the renewable energy so that the demand for the energy increases.• Fossil fuels raise the mortality rate (under 5).Fossil fuel is a non-renewable energy source that deteriorates the environment by emitting higher emissions and making the environment inhabitable.It also creates many unknown diseases in the human body and affects the infant vastly; therefore, child mortality under age 5 increases as a result of increasing fossil fuel consumption.So, the governments of the G-7 countries should ban nonrenewable energy sectors and transfer them to renewable energy sectors.Moreover, the government can also decrease the demand for fossil fuels by imposing a higher tax on them.• Industrialization enhances the under-5 child mortality rate.So, the governments of the G-7 countries should adopt green technology or environmentally friendly technologies in industry.Moreover, the government can make the use of renewable energy mandatory.In addition, the products or services that emit more should be banned immediately.Besides, the industries that break the rules or don't use green technology should be fined.
• As long as unemployment doesn't reduce the infant mortality rate.So, the government should ensure full employment so that people can properly take care of their children.Moreover, the government can also provide unemployment benefits so that they are capable of taking advantage of health services.In addition, the government can also increase the productivity of the workers to reduce unemployment.As a result, they will enjoy a better life, and it will significantly reduce the mortality rate.• As education increases, the mortality rates in the G-7 countries drop below 5. Child marriages and child mortality could be reduced if all women received a primary education.Education equality increases job opportunities and economic expansion, lowering the under-5 mortality rate.So, in order to reduce infant mortality in G-7 countries, policymakers should focus more on promoting the role of women by discarding institutional and cultural barriers that prevent women from accessing education.In addition, the G-7 government must provide education to disadvantaged girls and health counselling to women, particularly among scheduled castes or scheduled tribes, with a greater emphasis on backward regions in order to reduce mortality rates.

Table 2 .
This table demonstrates that the data set contains no unusual patterns.The mean statistics for all variables are consistent.The standard deviation The dependent variable, Infant Mortality (lnIMR), specifies the mortality rate per 1,000 live births, serving as a crucial indicator of a nation'

Table 1 .
List of variables

Table 2 .
Summary statistics

Table 3 .
Results of the CSD test Source: Author's estimation.

Table 4 .
Unit root tests
health output model (HO) is applied because, for this analysis, this model appropriately explains the theoretical framework of this paper.For properly selecting the long run estimation model, a number of diagnostic tests are conducted, such as the cross-sectional dependence test, the unit root test, and the cointegration test.Due to the presence of the cross-sectional dependence, unit root problem, and cointegration problem, this paper applied the Driscoll-