The study focuses on understanding the factors driving Total Factor Productivity (TFP) differences across emerging economies. Existing literature highlights the importance of human capital (health and education) in influencing TFP through various channels like technological innovation, labor market efficiency, and input productivity. Early research often employed aggregate measures like life expectancy and average years of schooling, which provided useful insights but lacked granular detail for effective policymaking. The World Health Organization (WHO) points out that life expectancy alone doesn't fully capture health's impact on productive capacity. Similarly, aggregate education indicators fail to represent diverse aspects of a country's educational system. The relationship between pollution and economic growth has been well-documented, although the relationship between pollution and TFP remains a key area of inquiry. While the impact of energy on growth is widely studied, its connection with TFP requires more exploration. Previous studies have examined the TFP-energy-pollution nexus but often lack a comprehensive model incorporating other relevant variables like various levels of education, undernourishment, water access, and HIV prevalence. These factors are crucial in determining human capital quality, along with environmental quality and information technology. Environmental quality impacts TFP through its effects on health. Information and Communication Technology (ICT) improves TFP indirectly by enhancing the quality of education and production. This study addresses these gaps by comparing aggregate and disaggregate measures of health and education, analyzing the impact of factors determining human capital quality (environmental quality and ICT), and examining the TFP-energy-pollution nexus within a broader economic model.

The literature review highlights the debate surrounding the inclusion of human capital in production functions or TFP equations. Some studies have found a positive correlation between human capital and TFP, particularly in developing countries with increased trade openness. However, others find human capital insignificant, especially in panel data studies. Existing research often relies on secondary schooling or average years of schooling, neglecting the varied impacts of different educational levels on productivity. Studies on primary, secondary, and tertiary education's impact on economic growth provide varied results, emphasizing the need to consider all levels to support stable development. In addition to health and education, variables like trade openness (knowledge transfer and technology transfer), energy consumption, ICT, and industrial share of GDP are often included in TFP models due to their suggested roles in the broader macroeconomic structure. Trade openness, especially, has been positively linked to TFP by several studies. Furthermore, the literature review underscores that macroeconomic factors like trade openness, energy consumption, ICT, and industrial share of GDP need to be included in a comprehensive TFP model. Previous works have investigated the effect of individual indicators but have not provided the holistic analysis presented in this paper.

The study employs a panel data approach, analyzing data from 30 emerging economies from 1990-2019. The choice between random effects and fixed effects models is determined using the Hausman test, which indicates that a fixed effects model is more suitable. Total factor productivity (TFP) data is sourced from the Penn World Table 10.0. Other variables such as Trade Openness, Energy Use Per Capita, Industrial Share to GDP, Information and Communications Technology (ICT) services, various measures of education (average years of schooling, primary, secondary, tertiary), life expectancy, HIV/AIDS prevalence, access to clean water, undernourishment and air pollution are sourced from the World Bank's World Development Indicators (2020) and Barro and Lee's Educational Attainment Dataset (2020). A Cobb-Douglas production function is not used directly; instead, the study focuses on the impact of health and education indicators on TFP. The study addresses potential multicollinearity issues by computing a correlation matrix (Table 2), showing that no pair of variables exhibits high correlation (r<0.9). Descriptive statistics (Table 3) present the mean, standard deviation, minimum, maximum, skewness, and kurtosis of all variables. The main empirical model is specified as: In tfp₁ = θ₀ + θ₁lnX + θ₂lntrad + θ₃ln ent + θ₄ln indus + θ₅ln ICT + εit, where tfp represents total factor productivity, X represents health and education indicators, trad represents trade openness, en represents energy consumption per capita, indus represents industrial share to GDP, and ICT represents information and communication technology services. Multiple fixed-effects models are estimated to analyze the impact of different combinations of the variables on TFP. The models are estimated to capture the impact of primary, secondary, and tertiary education separately. A model is estimated using average years of schooling.

The Hausman test results reject the null hypothesis of the random effects model being appropriate (p-value < 0.05), confirming the suitability of the fixed effects model for the analysis. Table 4 presents the estimated fixed-effects model results. The results across models consistently show positive and statistically significant coefficients for primary, secondary, and tertiary schooling, indicating a positive relationship between each level of education and TFP. Tertiary education exhibits a larger positive impact on TFP than secondary education. This difference may be due to specialization and skill-matching within the economy's industrial structure. It suggests a diminishing returns to scale for education; higher education may contribute less to productivity when there is a mismatch between skill supply and demand. The coefficients for energy use, industrial share, and ICT services also demonstrate a positive relationship with TFP. Trade openness consistently exhibits a negative and statistically significant relationship with TFP. In models incorporating disaggregate measures of health (undernourishment, lack of access to clean water, air pollution, and HIV prevalence), the study finds that undernourishment, lack of access to clean water, air pollution, and HIV prevalence have negative relationships with TFP, while life expectancy has a positive relationship with TFP. The study's findings emphasize the importance of human capital development and the need to address challenges like malnutrition, lack of access to clean water, and HIV/AIDS to improve TFP.

The findings highlight the significant role of health and education in boosting TFP in emerging economies. The positive and statistically significant coefficients for different educational levels emphasize the importance of investing in all levels of education, aligning skill development with industrial needs to enhance productivity. The negative effect of trade openness underscores the importance of policies that facilitate technology transfer and knowledge sharing, ensuring that economic integration benefits productivity. The offsetting effects of energy use and pollution highlight the need for sustainable energy strategies and policies that promote a cleaner energy mix. The negative impact of undernourishment, poor water access, air pollution and HIV/AIDS on TFP underlines the need for improved public health infrastructure and interventions to address these critical issues. The findings provide policy implications for improving human capital, particularly in the education sector, to drive TFP growth and sustainable economic development.

The study's main contribution lies in its comprehensive analysis of the combined impact of health, education, energy, and pollution on TFP in emerging economies. The findings underscore the critical role of human capital investment in fostering productivity and economic growth, emphasizing the need for targeted interventions to address health and education challenges. Further research could explore the dynamic interactions between these factors and investigate the long-term impacts of policies designed to improve human capital and environmental sustainability. Further research could explore specific policy interventions tailored to each identified challenge to assess their potential for enhancing TFP growth in different contexts.

The study's limitations include the potential for omitted variable bias, given the complexity of factors influencing TFP. The use of aggregate indicators for some variables might mask heterogeneity across the sample countries. The study focuses on a specific set of emerging economies and its findings may not be universally generalizable to all emerging economies. Further studies incorporating regional-specific analysis and exploring alternative model specifications could address these issues.