The impact of government spending on well-being: a case of upper middle-income countries and high-income countries

Gumede, Ruth T.; Greyling, Lorraine; Mazorodze, Brian T. Article The impact of government spending on well-being: a case of upper middle-income countries and high-income countries Cogent Economics & Finance Provided in Cooperation with: Taylor & Francis Group Suggested Citation: Gumede, Ruth T.; Greyling, Lorraine; Mazorodze, Brian T. (2024) : The impact of government spending on well-being: a case of upper middle-income countries and high-income countries, Cogent Economics & Finance, ISSN 2332-2039, Taylor & Francis, Abingdon, Vol. 12, Iss. 1, pp. 1-18, https://doi.org/10.1080/23322039.2024.2413657 This Version is available at: https://hdl.handle.net/10419/321627 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Cogent Economics & Finance ISSN: 2332-2039 (Online) Journal homepage: www.tandfonline.com/journals/oaef20 The impact of government spending on wellbeing: a case of upper middle-income countries and high-income countries Ruth T. Gumede, Lorraine Greyling & Brian T. Mazorodze To cite this article: Ruth T. Gumede, Lorraine Greyling & Brian T. Mazorodze (2024) The impact of government spending on well-being: a case of upper middle-income countries and high-income countries, Cogent Economics & Finance, 12:1, 2413657, DOI: 10.1080/23322039.2024.2413657 To link to this article: https://doi.org/10.1080/23322039.2024.2413657 © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group Published online: 13 Oct 2024. Submit your article to this journal Article views: 1376 View related articles View Crossmark data Citing articles: 1 View citing articles Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=oaef20 DEVELOPMENT ECONOMICS | RESEARCH ARTICLE The impact of government spending on well-being: a case of upper middle-income countries and high-income countries Ruth T. Gumede a , Lorraine Greyling a and Brian T. Mazorodze b a Department of Economics, Faculty of Commerce, Administration and Law, University of Zululand, Richards Bay, South Africa; b School of Economic and Management Sciences, Sol Plaatje University, Kimberley, South Africa ABSTRACT The welfare effects of government redistributive policy have been subject to considerable debate for decades. Against this background, this study explores the effect of government social spending on an empirically constructed measure of well-being in a panel of 16 upper middle-income countries and 38 high-income countries observed between 2002 and 2019. The study utilises the Generalised Method of Moments (GMM) Model to estimate the empirical relationship between government social spending and well-being. The results suggest that welfare gains in upper middle-income countries are derived from redistributive spending that prioritises schooling. On the other hand, rich countries are more likely to benefit from health-related spending. Based on the results the study confirms that disaggregated social spending in upper middle-income nations and wealthy nations does not impact aggregate well-being uniformly. Therefore, efforts to improve aggregate welfare through government redistributive spending ought to consider these attendant heterogeneities. IMPACT STATEMENT This study examines the association between government social spending and wellbeing using a panel dataset of 54 upper middle-income and high-income countries. The novel feature of this investigation is embedded in the construction of the aggregate well-being index. We employed the Principal Components Analysis (PCA) technique to build a composite well-being index. This well-being indicator incorporates age dependency, access to water, access to sanitation, and life expectancy, the institutional quality is measured by four indicators which are government effectiveness, control of corruption, political stability and rule of law, environment degradation (CO2 emissions) and economic growth (GDP). This approach is expected to contribute substantially to the scarce literature in the field of well-being in upper middle-income countries and higher-income countries. A focus on well-being research is important since improved well-being leads to a better quality of life and stimulates economic performance through its influence on human development. The findings suggest that welfare gains in upper middle-income economies are attributed to social spending that prioritizes education. In contrast, wealthy countries are more likely to benefit from health-related spending. Policymakers should develop redistributive policies that address the specific needs of each region since evidence demonstrates that disaggregated social spending in upper middle-income countries and wealthy nations does not influence aggregate well-being uniformly. ARTICLE HISTORY Received 30 June 2024 Revised 20 September 2024 Accepted 2 October 2024 KEYWORDS Well-being; government spending; education; health; System GMM SUBJECTS Economics; Economics and Development; Political Economy JEL CODES I18; I25; I31 1. Introduction Fiscal policy is part of the toolkit that governments employ to achieve their macroeconomic objectives (Mankiw, 2010). In the Keynesian school of thought, fiscal policy can stabilise cyclical fluctuations in output and employment in the short run. This understanding has motivated a strand of empirical research probing the economic effects of fiscal policy (Chipaumire et al., 2014; Chude & Chude, 2013; Odhiambo, CONTACT Ruth T. Gumede [email protected],[email protected] Department of Economics, Faculty of Commerce, Administration and Law, University of Zululand, Private Bag X1001, KwaDlangezwa 3886, Richards Bay, South Africa. ß2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. COGENT ECONOMICS & FINANCE 2024, VOL. 12, NO. 1, 2413657 https://doi.org/10.1080/23322039.2024.2413657 2015). However, fiscal policy also facilitates the redistribution of income. The primary objective of redistributive policies is to advance the welfare of the society. Redistributive programs advocate for equality by overseeing the transfer of resources from the wealthy class to the less privileged people. Therefore, building on the foundation of the egalitarian school of thought which encourages equality among people. The concept of well-being can be observed from two different perspectives, objective and subjective viewpoints. The subjective approach deals with non-economic characteristics of well-being. The subjective approach describes people’s assessment of their lives, especially their contentment with their lives. In contrast, objective well-being captures material resources that significantly determine the basic needs of households, such as distribution of income, household consumption, employment, expenditures etc. The well-being concept is essential and requires more attention since improved well-being not only promotes the quality of life but also stimulates economic growth through the human development channel. In general, people with improved well-being tend to progress through the education system more smoothly and highly productive at work than compared to those with poor health and education outcomes. The emergence of the endogenous growth paradigm has highlighted the significance of the redistributive components of fiscal policy, which have been undervalued in the literature. According to this theory, public investment in the social sectors (such as health and education) promotes human development which is a significant factor in sustainable growth and well-being (Lucas, 1988; Romer, 1990). In addition, the formulation of the Millennium Development Goals (MDG) in 2000 also reinforced the importance of social policy in eradicating extreme poverty and advancing well-being (United Nations, 2015). There is a dearth of welfare studies, the majority of which focus on the quality of education (years of schooling and school enrolment), health (life expectancy, mortality), income redistribution (Gini coefficient), and economic growth (real GDP or GDP per capita) as indicators for aggregate welfare (Anderson et al., 2017; Arthur & Oaikhenan, 2017; Craigwell et al., 2012; Karim, 2015; Ndaguba & Hlotywa, 2021). A key challenge facing these studies is the complexity surrounding the definition and measurement of well-being. Recently, a troubling pattern has emerged where economic growth is associated with persistent inequality and socio-economic issues, causing economists to challenge the notion that economic growth is the appropriate indicator of aggregate well-being. Consequently, there are growing demands for the development of new measurements which embody holistic well-being. To address the knowledge gap, this study explores the association between government social spending and welfare using a panel dataset of 54 upper middle-income and high-income countries. The findings will contribute to the scarce literature in terms of redistribution policy and well-being. The novel feature of this investigation is embedded in the development of the well-being index. To achieve this uniqueness, the study will use the Principal Components Analysis (PCA) technique to build a composite well-being index. The PCA technique integrates several characteristics into a single indicator, thus allowing for the inclusion of several significant well-being variables in a single regression. Furthermore, utilising disaggregated government spending on education, health and social protection has the advantage of facilitating your targeted fiscal policymaking in terms of discretionary decisions. The remainder of the paper is arranged as follows: Section 2 focuses on the literature review, Section 3 documents methodology (which covers data description and estimation strategy), and empirical results and discussion are provided in Section 4. The robustness findings are reported in Section 5,whiletheconclusion and policy recommendation are presented in Section 6. 2. Literature review For decades, economists and policymakers have focused on GDP as a composite indicator of well-being. However, the notion that ‘income (GDP) is a better indicator for well-being’has recently been contested by several economists, arguing that GDP is simply a subset of the complex concept of well-being (Aitken, 2019; Decancq & Schokkaert, 2016; Stiglitz et al., 2019). The Nobel laureate Amartya Sen’s earlier research played a major role in the development of the well-being indicator. Sen (1985) thus suggested that when economic growth indicator like GDP is utilised as a proxy for aggregate well-being people’s freedoms are neglected. According to this framework, well-being can be characterised as a multidimensional domain that looks at life quality and considers non-financial aspects including leisure activities, social interactions, 2 R. T. GUMEDE ET AL. and the standard of education and health. Sen’s(1985) study led to the formation of the Human Development Index (HDI) under the United Nations Development Programme (UNDP) in 1990. The HDI measured life expectancy at birth, GDP per capita, and educational attainment. According to the United Nations (UN), the three factors promote aggregate well-being (Shrotryia & Singh, 2020). On the other hand, Stiglitz et al. (2019), Bro cek and Lalinsk y(2017) and Jones and Klenow (2016) also analysed the concept of beyond GDP. They argued that employing the wrong indicator for example, GDP as a proxy for aggregate well-being could distort the truth of how people are faring in the country and induce the government to implement irrelevant policies. According to these researchers’well-being studies must incorporate income distribution, the quality of education, health-related indicators, environmental changes, and all factors influencing the standard of living. Finally, scholars like Wiseman and Brasher (2008)and Atkinson et al. (2020) have developed a conceptual framework for community well-being recently. Accordingly, community well-being is determined by multiple domains (social, economic, political, cultural, environmental, etc.). These factors assist individuals and communities achieve their goals through building strong social networks of relationships and promoting liveable and equitable communities. When exploring the well-being and the redistributive aspects of government policy it is imperative to discuss the egalitarian school which seeks to address the inequalities within the society through the distribution of resources, and wealth and providing equal opportunities to all citizens. Traditionally egalitarianism framework suggested that the government should diminish the inequalities originating from external factors while ignoring the inequalities created by people’s decisions (Bengtson & LippertRasmussen, 2023; Dworkin, 1981). This notion was opposed by some egalitarians who claimed that people who have made bad decisions a left to suffer, thus antithetical to the egalitarian perspective whose primary objective is to promote equality for all citizens. Egalitarianism should place more emphasis on relational equality rather than merely equalising people’s circumstances. Relational equality proposes that in an egalitarian society, everyone is awarded the same fundamental rights and receives equal treatment. Those who advocate for relational egalitarianism believe that the equality concept is best explained by the relationships that exist among people rather than the distribution of resources (Anderson, 1999; Scheffler, 2010). In a nutshell, the egalitarianism ideology aims to promote equality among people, which can be achieved through improving the conditions of the less privileged, creating equal economic opportunities, and treating everyone equally. Meltzer and Richard (1981) presented a majority-voting model, where a higher level of income inequality leads to higher redistribution efforts. Most people tend to prefer a government that redistributes capital from the wealthy group to the underprivileged group when the population earning mean income is greater than those earning the median income (Gumede et al., 2019). The Meltzer and Richard model is the extension of the median voter theorem founded by Downs (1957). This suggests that in a democratic country which adopts the majority rule, the voting process will choose the outcome that is most popular with the median voter. The model is focused on two assumptions, The first one is that democratic regime voters can categorise all options into a single political spectrum. Secondly, the voter’s preferences are single-peaked. Overall, the preferences of a median voter determine the size of the government expenditures and taxes. From an empirical standpoint, Haile and Ni~ no-Zaraz ua (2018) sought to investigate whether social spending stimulates welfare in 55 low-income and middle-income economies with panel data spanning from 1990 to 2009. Social spending was denoted by education, health, and social protection spending, while aggregate welfare was proxied by the inequality-adjusted human development index (IHDI), human development index (HDI), and child mortality rates. Their results indicated that aggregate social spending stimulates welfare in both low-income and middle-income countries. In the case of disaggregated social spending, all components have a positive sign as expected, but only the health spending estimate promotes welfare. From a time series approach, Ndaguba and Hlotywa (2021) evaluated the relationship between government health expenditure and well-being. The ARDL estimates confirm that health spending is positively related to the human development index (HDI). Thus, investment in public health stimulates well-being in South Africa. In another study, Craigwell et al. (2012) examined the impact of government education and health spending on economic growth and the quality of life using panel data from 19 Caribbean countries. The findings indicated that health spending boosted life expectancy, while education spending had a significant impact on primary and secondary school enrolment. In addition to the above studies. COGENT ECONOMICS & FINANCE 3 Arthur and Oaikhenan (2017) empirically assessed the influence of government expenditure on wellbeing in sub-Saharan Africa. They found that raising health spending prolonged life expectancy and mitigates mortality rate. The study by Niehues (2010) evaluated the effect of government redistribution expenditure on income inequality utilising panel data from 24 European countries. The GMM findings suggested that social programmes dealing with unemployment benefits and social pensions narrow the gap between the wealthy and the underprivileged. The welfare study conducted by Ospina (2010) revealed that public spending on education and health promotes equality in Latin America. However, social security spending promotes inequality. Samir Saad (2024) panel research explored the role of public spending on inclusive growth. Through the principal component analysis (PCA) the study computed the composite inclusive growth index which took into account factors such as quality education and health, environment sustainability, and income distribution. Based on the findings, government intervention through education and health spending raised inclusive growth. A study by Amaluddin et al. (2018) sought to determine whether a modified human development index (HDI) constructed using the PCA significantly influenced well-being in 33 Indonesian villages. Following the United Nations methodology, the modified HDI indicator emerged years of schooling, income per capita and life expectancy, while the poverty rate served as the dependent variable. The outcome revealed that as human development advances, poverty declines, resulting in improved well-being in Indonesia. Based on the empirical section, it appears limited studies have explored the link between government disaggregated social spending and aggregate well-being, most of which do not treat well-being as a multidimensional concept. Secondly, the mentioned studies have neglected institutional quality, although quality institutions play a significant role in well-being and social policy. The establishment of sound institutions stimulates inclusive growth and well-being through reducing transaction costs, protecting human rights, honouring contractual agreements, ensuring governments operate effectively and finally, minimising fraudulence activities. Thirdly, environmental factors have been overlooked in welfare studies. In recent years climate change has become more aggressive, causing severe damage to the environment and human well-being. Researchers must therefore take environmental sustainability into account when studying well-being. To address the mentioned issues, we employ the Principal Components Analysis (PCA) technique to build a composite well-being index. This well-being indicator incorporates age dependency, access to water, access to sanitation, and life expectancy, the institutional quality is measured by four indicators which are government effectiveness, control of corruption, political stability and rule of law, environment degradation (CO 2 emissions) and economic growth (GDP). The applied methodology will contribute significantly to the sparse literature in the field of well-being studies in upper middle-income countries and higher-income countries. 3. Methodology To analyse the empirical association between government social spending and aggregate well-being we employed a panel dataset consisting of 54 countries spanning from 2002 to 2019. The sampled economies are categorised into 16 upper middle-income countries and 38 high-income economies following the World Bank classification. Furthermore, the selection of the countries and the sampling period were dictated by data availability. The list of selected nations and the data description are reported in the Appendix section, Tables A1 and A2, respectively. The secondary data was extracted from the OECD Social Expenditure Database (SOCX), IMF Government Finance Statistics (IMF-GFS), World Development Indicators (WDI), Economic Commission for Latin American and the Caribbean Countries (CEPALSTAT), Worldwide Governance Indicators (WGI) and the Eurostat Database. 3.1. Welfare empirical model Guided by empirical literature (Haile & Ni~ no-Zaraz ua, 2018), we estimated different variants of the following dynamic 1 model. Wi,t¼ h 0Wi,t−1þb1Ei,tþb2Hi,tþb3Si,tþcXþgiþttþei,t(1) 4 R. T. GUMEDE ET AL. i¼1, ...,n;t¼1, ...,T From equation (1) Wi,tdenotes well-being; Wi,t−1represents one period lagged well-being, while i indexes individual country and ttime period (year). b1,b2and b3are the main parameters of inquiry, measuring the impact of disaggregated social spending (education, health and social protection, respectively) on aggregate well-being, On the other hand, Xis a vector of other variables that are likely correlated with wellbeing (Wi,t) and predictor variables (Ei,t,Hi,tand Si,t). The selected control variables are employment in service sectors (% of total employment), gross capital formation (GCF) and inflation, Finally, giand ttstands for country-specific fixed effects and time-fixed effects respectively, while the ei,tdepicts the error term. Our well-being index incorporates age dependency, access to water, access to sanitation, and life expectancy, the institutional quality is measured by four indicators which are government effectiveness, control of corruption, political stability and rule of law, environment (CO 2 emissions) and economic growth (GDP). Noteworthy, the concept of welI-being is very comprehensive and thus measured by several indicators. However, it is impossible to incorporate all well-being variables into a single regression model. Furthermore, most of the well-being indicators are highly correlated (Haile & Ni~ no-Zaraz ua, 2018). Consequently, we employ the principal component analysis (PCA) approach to combine the selected well-being variables into a single variable. This process allows the replacement of huge numbers of highly correlated well-being indicators with small numbers of uncorrelated indicators. Through the consolidation of several variables into a single indicator, PCA replicates the original dataset. The age dependency ratio is a crucial factor in assessing well-being and economic growth. As the age dependency ratio rises, savings and investments decline, which in turn lowers economic growth and living standards. Conversely, lower dependency ratios promote economic performance by improving overall savings and investment (Oliver, 2015). Overall, the population’s structure affects economic growth and well-being, making the age dependency ratio an important indicator of aggregate well-being. Having access to improved water and sanitation facilities is essential for human survival. Quality water and sanitation services that are managed safely stimulate people’s hygiene and health outcomes, which improves their overall well-being (Armah et al., 2018; Pullan et al., 2014). Conversely, tainted water and inadequate sanitation facilities exacerbate illnesses like cholera, hepatitis, diarrhoea, typhoid, and dysentery, impairing health and well-being. Another variable included in the formation of the well-being index is the life expectancy metric. Accordingly, higher life expectancy typically signifies improved health and an adequate standard of living for individuals. Improved life expectancy fosters human development, which is normally used as an indicator of well-being (Haile & Ni~ no-Zaraz ua, 2018). On the other hand, institutional quality is a significant determinant of both economic development and aggregate welfare. Sound institutions foster sustainable growth by reducing costs associated with production, transactions and manufacturing (North, 1990). Institutional quality drives the government to perform its functions effectively and efficiently (Le & Kim, 2021; Sabir et al., 2019). Therefore, in the computation of the well-being indicator, we include the following variables to account for institutional quality: political stability, rule of law, government effectiveness, and control of corruption. The detrimental impacts of climate change that have persisted over the past few decades have spurred conversations about practical strategies for reducing greenhouse gas emissions. The constant increase in greenhouse gas emissions has triggered severe weather phenomena like floods, droughts, elevated temperatures, and the deterioration of ecosystems. Consequently, greenhouse gas emissions are harmful to human well-being and the environment (Omri, 2013). A previous study by Majeed and Ozturk (2020) found that higher emissions of carbon dioxide (CO2) lead to lower life expectancy and higher infant mortality rates. After considering the significant influence greenhouse gas emissions have on health outcomes, we choose to incorporate this pollutant variable into our well-being index. 0 Finally, we incorporate real GDP as a proxy for economic growth, into our well-being index. Economic growth is essential to people’s well-being since it constitutes a substantial portion of objective wellbeing (Jones & Klenow, 2016). Overall, every variable that went into creating our well-being indicator is essential to human development, and our study is anticipated to make a significant contribution to the scant literature on well-being. Disaggregated social spending will be based on three components social COGENT ECONOMICS & FINANCE 5 protection, health and education. Consequently, the work of Haile and Ni~ no-Zaraz ua (2018), Lustig (2018), and Khan and Bashar (2015) guided the selection of these variables. The control variables well-being model will include employment, investment and inflation. The growth of employment stimulates economic performance and the standard of living (Atkinson, 2015). Employed individuals can enhance their quality of life without transferring the financial burden to the government (Dev, 2018). On the other hand, inflation is the most important macroeconomic variable, and even little changes in this indicator could negatively impact people’s quality of life. Given that rising prices imply declining consumer purchasing power, high inflation is generally linked to poor well-being (Yolanda, 2017). Finally, investment is fundamental to sustainable growth because it simultaneously increases aggregate supply and demand and has positive externalities on well-being (Mankiw, 2010). Consequently, the study employs gross capital formation (GCF) as a proxy for domestic investment. 3.2. Principal component analysis framework To analyse the influence of government social spending on aggregate welfare, we construct a well-being index using the principal component analysis (PCA). The principal components analysis is generally portrayed as a multivariate approach that is responsible for reducing several variables into smaller orthogonal principal components, through linear combinations of those variables (Kulo glu & Ecevit, 2017; Sabir et al., 2019). Each new principal component incorporates as much information as possible that is not accounted for by previous components. Therefore, the initial component is responsible for the highest possible variation in the original variables. A critical feature of the structure is that the initial and second components are uncorrelated. Furthermore, the third component captures maximum information which is not presented in the first and second components. The P variables of the model depict the structure interdependence and are also converted into new variables incorporating properties of linear, orthogonal and uncorrelated (Dalton-Greyling & Tregenna, 2014). 3.3. Econometric strategy Traditionally a panel estimation process uses unit root tests to evaluate the stationarity properties of the dataset to determine the suitable estimation strategy. However, Kitamura and Phillips (1997), Sarpong and Nketiah-Amponsah (2022) and Okafor et al. (2015) stated that a non-stationary series can generate robust and reliable outcomes under the conditions that the sample size is short, and the number of entities (N) exceeds the time period (T). The serial correlation issue is mitigated when these prerequisites are met. Furthermore, there is a possibility that our welfare empirical analysis may be exposed to the endogeneity, meaning that social spending variables are not exogenously determined. Accordingly, government spending on social sectors (education, health and social protection) and well-being could be determined by the same unobserved factors. Also, well-being may influence social spending (a phenomenon referred to as reversed causality. Notably, if the endogeneity problem is not accounted for in the regression, then the generated slope of coefficients may be biased. Thus, to address the issue of endogeneity we employ the Generalised Method of Moments (GMM) estimator. 3.4. Generalised Method of Moments (GMM) The Generalised Method of Moments (GMM) framework was founded by Arellano and Bond (1991). The core properties of this dynamic estimator are to control for unobserved country heterogeneity, endogeneity issues, measurement errors and omitted variable bias problems. GMM generates robust results when dealing with small sample sizes and more cross-sectional observation. Furthermore, a GMM is considered to be superior to an instrument variable (IV) since, in the presence of heteroskedasticity, a GMM estimator is more efficient than an IV estimator (Haile & Ni~ no-Zaraz ua, 2018), whereas, in the absence of heteroscedasticity, the GMM estimator is no worse asymptotical than the IV estimator. The original GMM estimator is referred to as difference GMM and is expressed as follows: 6 R. T. GUMEDE ET AL. Wit −Wit−1¼a−1 ðÞ Wit þb0Xit þliþei,t(2) From equation 2, the left-hand side measures the growth rate of well-being at time (i), While Wit is the well-being index, iindexes individual country and ttime period (year), the Xit stands for a vector of explanatory variables, while the unobserved country-specific effect is denoted by li, last eistand for disturbance term. Rewriting equation (2), translate to: Wit ¼aWit−1þb0Xit þliþei,t(3) To eliminate country-specific effects, we take the first differences of equation 3: Wit −Wit−1¼aWit−1−Wit−2 ðÞ þb0Xit−Xit−1 ðÞþ eit −eit−1(4) The GMM panel estimator uses the following moment conditions: EW it −seit −eit−1 ðÞ  ¼0 for s 2; t ¼3, …,T EX it −seit −eit−1 ðÞ  ¼0 for s 2; t ¼3, …,T The difference GMM estimator utilises the initial values of explanatory variables and the lagged for dependent variables as instruments. To generate robust and reliable coefficients under the first difference specification the hypotheses of weak exogenous variables and the absence of second serial correlation in disturbance terms must hold. However, like other econometric modelling approaches the first difference GMM estimator has some limitations. Firstly, when the regressors are persistent over time, their lagged levels become poor instruments for the variables in the different specifications. Secondly, the issue of weak instruments also prevails when the sample size is short (Arellano & Bover, 1995; Blundell & Bond, 1998; Kumar, 2013). Blundell and Bond (1998) introduced the extension of different GMM which is referred to as the system GMM estimator. This new estimation approach constructs the system GMM using two equations. The original equation (first difference equation with levels as instruments) and additional equation expressed in levels with first differences as instruments. Therefore, the moment of conditions for the added equation in levels is depicted as follows: EW it−s−Wit−s−1 ðÞ li−ei,t ðÞ  ¼0 for s ¼1 EX it−s−Xit−s−1 ðÞ li−ei,t ðÞ  ¼0 for s ¼1 Blundell and Bond (1998) concluded through the Monte Carlo simulations that the extended GMM model is comparatively more efficient than the first-difference GMM model. Therefore, we employ the system GMM as the baseline estimator. 3.5. Robustness testing: Panel Corrected Standard Errors (PCSE) The system GMM estimator addresses endogeneity issues. However, the problem of contemporaneous correlation, which arises when the idiosyncratic disturbance factors are connected across countries, is ignored by the system GMM. As a result of trade links, there is a strong likelihood that cross-sectional dependency may arise among the observed nations. Therefore, to conduct our robustness assessment we employ the Panel Corrected Standard Error (PCSE) estimator pioneered by Beck and Katz (1995). This sophisticated technique accounts for cross-sectional dependence, heteroscedasticity, and autocorrelation. The PCSE estimator yields robust estimates when the number of cross-sections (N) is greater than the time dimension (T). Given that the cross-sectional units (54 countries) exceed the time dimension (2002– 2019), the PCSE model is therefore appropriate for our situation (Bailey & Katz, 2011). 4. Results and discussion This section provides an empirical analysis of the nexus between government social spending and aggregate well-being in upper middle-income and high-income economies. It is crucial to ascertain whether there are collinearity problems among the explanatory variables in each model before presenting our estimation results. The results of the pairwise correlation matrix test are shown in Table 1. Concerns for perfect collinearity between the explanatory variables are diminished in our situation because correlation COGENT ECONOMICS & FINANCE 7 should primarily focus on the health sector. The proposed implementation should occur without abandoning the other redistributive programs. 6.2. Limitation and future research During the construction of our welfare index, we concentrated primarily on objective well-being factors, while overlooking subjective well-being. Future studies should use mixed methods since the qualitative side of this approach will allow researchers to capture the subjective aspects of well-being which relate to human psychology. Therefore acquiring subjective well-being data through interviews, questionnaires and surveys. On the other hand, the quantitative side will focus on objective well-being. There will be significant contributions to the field of well-being resulting from the application of this methodology. Note 1. A dynamic model is preferred over a static model to capture inertia effects and the persistent nature of welfare improvements. Acknowledgements We are grateful to Professor I. Kaseeram from the University of Zululand, Department of Economics for knowledge sharing and suggestions during the initial phases of our study. Authors contributions Conceptualization, R.T.G; B.T.M; and L.G; Methodology, R.T.G.; B.T.M; Software, R.T.G.; B.T.M; Formal analysis, R.T.G.; Resources, R.T.G.; Data curation, R.T.G.; Writing—original draft, R.T.G.; Supervision, L.G. and B.T.M. All authors have read and agreed to the published version of the manuscript. Disclosure statement No potential conflict of interest was reported by the author(s). Funding This project was sponsored by the National Research Foundation bursary scheme, unique Grant No: 118433. About the authors Ruth Thandazile Gumede is a doctoral student (PhD) at the University of Zululand, Department of Economics in South Africa. Her research interests include Well-being, Social policy, Political Economy and development economics. Lorraine Greyling is the Dean of the Faculty of Commerce, Administration, and Law, and professor of Economics at the University of Zululand in South Africa, She holds a doctorate in Economics. Her research interests are Macroeconomics, Development Policy Issues, Econometrics, Economic History and Quantitative Analysis. She is the chair of the Institutional Forum and is a member of the Council of the University of Zululand. Brian Tavonga Mazorodze holds a doctorate in Economics and lectures undergraduate courses in Statistics and Econometrics at Sol Plaatje University South Africa. His area of specialisation includes Trade, Development and Industrial economics. ORCID Brian T. Mazorodze http://orcid.org/0000-0002-7799-4627 Data availability statement The study’s dataset is available upon request from the corresponding author, R.T. Gumede. 14 R. T. GUMEDE ET AL. References Aitken, A. (2019). Measuring welfare beyond GDP. National Institute Economic Review,249,R3–R16. https://doi.org/10. 1177/002795011924900110 Amaluddin, A., Payapo, R. W., Laitupa, A. A., & Serang, M. R. (2018). A modified human development index and poverty in the villages of west seram regency, Maluku province, Indonesia. International Journal of Economics and Financial Issues,8(2), 325–330. Anderson, E. S. (1999). What is the point of equality? Ethics,109(2), 287–337. https://doi.org/10.1086/233897 Anderson, E., Jalles D’Orey, M. A., Duvendack, M., & Esposito, L. (2017). Does government spending affect income inequality? A meta-regression analysis. Journal of Economic Surveys,31(4), 961–987. https://doi.org/10.1111/joes. 12173 Arellano, M., & Bond, S. (1991). Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. The Review of Economic Studies,58(2), 277–297. https://doi.org/10.2307/2297968 Arellano, M., & Bover, O. (1995). Another look at the instrumental variable estimation of error-components models. Journal of Econometrics,68(1), 29–51. https://doi.org/10.1016/0304-4076(94)01642-D Armah, F. A., Ekumah, B., Yawson, D. O., Odoi, J. O., Afitiri, A. R., & Nyieku, F. E. (2018). Access to improved water and sanitation in sub-Saharan Africa in a quarter century. Heliyon,4(11), e00931. https://doi.org/10.1016/j.heliyon. 2018.e00931 Arthur, E., & Oaikhenan, H. E. (2017). The Effects of health expenditure on health outcomes in Sub-Saharan Africa (SSA). African Development Review,29(3), 524–536. https://doi.org/10.1111/1467-8268.12287 Atkinson, A. B. (2015). Inequality: What can be done? Harvard University Press. Atkinson, S., Bagnall, A. M., Corcoran, R., South, J., & Curtis, S. (2020). Being well together: Individual subjective and community wellbeing. Journal of Happiness Studies,21(5), 1903–1921. https://doi.org/10.1007/s10902-019-00146-2 Aydan, S., Bayin Donar, G., & Arikan, C. (2022). Impacts of economic freedom, health, and social expenditures on well-being measured by the better life index in OECD countries. Social Work in Public Health,37(5), 435–447. https://doi.org/10.1080/19371918.2021.2018083 Bailey, D., & Katz, J. N. (2011). Implementing panel-corrected standard errors in R: The pcse package. Journal of Statistical Software,42(1), 1–11. https://doi.org/10.18637/jss.v042.c01 Beck, N., & Katz, J. N. (1995). What to do (and not to do) with time-series cross-section data. American Political Science Review,89(3), 634–647. https://doi.org/10.2307/2082979 Bein, M. A., Unlucan, D., Olowu, G., & Kalifa, W. (2017). Healthcare spending and health outcomes: evidence from selected East African countries. African Health Sciences,17(1), 247–254. https://doi.org/10.4314/ahs.v17i1.30 Bengtson, A., & Lippert-Rasmussen, K. (2023). Relational egalitarianism and moral unequals. Journal of Political Philosophy,31(4), 387–410. https://doi.org/10.1111/jopp.12299 Blundell, R., & Bond, S. (1998). Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics,87(1), 115–143. https://doi.org/10.1016/S0304-4076(98)00009-8 Bro cek, F., & Lalinsk y, T. (2017). Welfare in Slovakia and the EU-an alternative to GDP per capita. Biatec Journal of Central Banking,17(6), 17–23. Chipaumire, G., Ngirande, H., Method, M., & Ruswa, Y. (2014). The impact of government spending on economic growth: Case South Africa. Mediterranean Journal of Social Science,5(1), 109–118. https://doi.org/10.5901/mjss. 2014.v5n1 Chude, N. P., & Chude, D. (2013). Impact of government expenditure on economic growth in Nigeria. International Journal of Business and Management Review,1(4), 64–71. https://doi.org/10.1007/s13398-014-0173-7.2 Craigwell, R., Bynoe, D., & Lowe, S. (2012). The effectiveness of government expenditure on education and health care in the Caribbean. International Journal of Development Issues,11(1), 4–18. https://doi.org/10.1108/ 14468951211213831 Dalton-Greyling, T., & Tregenna, F. (2014). Construction and analysis of composite quality of life index for a region of South Africa. ERSA (Economic Research Southern Africa) Working Paper, 481. Decancq, K., & Schokkaert, E. (2016). Beyond GDP: Using equivalent incomes to measure well-being in Europe. Social Indicators Research,126(1), 21–55. https://doi.org/10.1007/s11205-015-0885-x Downs, A. (1957). An economic theory of political action in a democracy. Journal of Political Economy,65(2), 135– 150. https://doi.org/10.1086/257897 Dworkin, R. (1981). What is equality? Part 2: Equality of resources. Philosophy and Public Affairs,10, 283–345. Economic Commission for Latin American and the Caribbean Countries. (2021). https://www.cepal.org/en Eggoh, J., Houeninvo, H., & Sossou, G. A. (2015). Education, health and economic growth in African countries. Journal of Economic Development,40(1), 93–112. Eurostat Database. (2021). https://ec.europa.eu/eurostat/data/database Gumede,R.T.,Bulagi,B.M.,&Greling,L.(2019).The nexus between social spending and economic growth in South Africa: A cointegration approach [Master’s theses, University of Zululand]. University of Zululand e-theses repository. https://uzspace.unizulu.ac.za/server/api/core/bitstreams/77d41713-11df-4958-896e-eefa7d4af34a/ content COGENT ECONOMICS & FINANCE 15 Haile, F., & Ni~ no-Zaraz ua, M. (2018). Does social spending improve welfare in low-income and middle-income countries? Journal of International Development,30(3), 367–398. https://doi.org/10.1002/jid.3326 International Monetary Fund: Government finance statistic. (2021). https://data.imf.org/?sk=a0867067-d23c-4ebcad23-d3b015045405 Jones, C. I., & Klenow, P. J. (2016). Beyond GDP? Welfare across countries and time. American Economic Review, 106(9), 2426–2457. https://doi.org/10.1257/aer.20110236 Karim, M. R. (2015). Public education spending and income inequality. International Journal of Social Science and Humanity,5(1), 75–79. https://doi.org/10.7763/IJSSH.2015.V5.425 Khan, H., & Bashar, O. K. (2015). Social expenditure and economic growth: Evidence from Japan using cointegration and causality tests. The Journal of Developing Areas,49(4), 285–300. https://doi.org/10.1353/jda.2015.0137 Kitamura, Y., & Phillips, P. C. (1997). Fully modified IV, GIVE and GMM estimation with possibly non-stationary regressors and instruments. Journal of Econometrics,80(1), 85–123. https://doi.org/10.1016/S03044076(97)00004-3 Kulo glu, A., & Ecevit, E. (2017). The relationship between health development index and financial development index: Evidence from high-income countries. Ekonomi, Politika & Finans Aras¸tırmalarıDergisi,2(2), 83–95. https:// doi.org/10.30784/epfad.346935 Kumar, S. (2013). Systems GMM estimates of the health care spending and GDP relationship: a note. The European Journal of Health Economics,14(3), 503–506. https://doi.org/10.1007/s10198-012-0394-7 Le, A. H., & Kim, T. (2021). The impact of institutional quality on FDI inflows: the evidence from the capital outflow of Asian economies. The Journal of Asian Finance, Economics and Business,8(8), 335–343. Lucas, R. E. (1988). On the mechanics of economic development. Journal of Monetary Economics,22(1), 3–42. https:// doi.org/10.1016/0304-3932(88)90168-7 Lustig, N. (2018). Fiscal policy, income redistribution, and poverty reduction in lowand middle-income countries. Journal of Globalization and Development,7(1), 17–60. Dev, S.M. (2018). Inequality, employment and public policy. The Indian Journal of Labour Economics,61,1–42. Majeed, M. T., & Ozturk, I. (2020). Environmental degradation and population health outcomes: a global panel data analysis. Environmental Science and Pollution Research International,27(13), 15901–15911. https://doi.org/10.1007/ s11356-020-08167-8 Mankiw, N. (2010). Macroeconomics (7th ed.). Worth. Meltzer, A. H., & Richard, S. F. (1981). A rational theory of the size of government. Journal of Political Economy,89(5), 914–927. https://doi.org/10.1086/261013 Ndaguba, E. A., & Hlotywa, A. (2021). Public health expenditure and economic development: The case of South Africa between 1996 and 2016. Cogent Economics & Finance,9(1), 1905932. https://doi.org/10.1080/23322039.2021. 1905932 Niehues, J. (2010). Social spending generosity and income inequality: a dynamic panel approach. The institution of the study of labour. IZA Working Paper No 5178. https://doi.org/10.2139/ssrn.1724733 North, D. C. (1990). Institutions, institutional change and economic performance. Cambridge University Press. Odhiambo, N. M. (2015). Government expenditure and economic growth in South Africa: an Empirical Investigation. Atlantic Economic Journal,43(3), 393–406. https://doi.org/10.1007/s11293-015-9466-2 Okafor, G., Piesse, J., & Webster, A. (2015). The motives for inward FDI into Sub-Saharan African countries. Journal of Policy Modeling,37(5), 875–890. https://doi.org/10.1016/j.jpolmod.2015.07.003 Oliver, M. (2015). Population ageing and economic growth in Japan. International Journal of Sociology and Social Policy,35(11–12), 841–863. https://doi.org/10.1108/IJSSP-02-2015-0018 Omri, A. (2013). CO2 emissions, energy consumption and economic growth nexus in MENA countries: Evidence from simultaneous equations models. Energy Economics,40, 657–664. https://doi.org/10.1016/j.eneco.2013.09.003 Organisation for Economic Co-operation and Development, Social Expenditure Database. (2023). https://www.oecd. org/social/expenditure.htm Ospina, M. (2010). The effect of social spending on income inequality: An analysis for Latin American countries. Center for Research in Economics and Finance (CIEF), Working Papers, (10–03). Popova, D. (2023). Impact of equity in social protection spending on income poverty and inequality. Social Indicators Research,169(1-2), 697–721. https://doi.org/10.1007/s11205-023-03167-w Pullan, R. L., Freeman, M. C., Gething, P. W., & Brooker, S. J. (2014). Geographical inequalities in use of improved drinking water supply and sanitation across sub-Saharan Africa: mapping and spatial analysis of cross-sectional survey data. PLoS Medicine,11(4), e1001626. https://doi.org/10.1371/journal.pmed.1001626 Romer, P. M. (1990). Endogenous technological change. Journal of Political Economy,98(5), S71–S102. https://doi.org/ 10.1086/261725 Sabir, S., Rafique, A., & Abbas, K. (2019). Institutions and FDI: evidence from developed and developing countries. Financial Innovation,5(1), 1–20. https://doi.org/10.1186/s40854-019-0123-7 Samir Saad, S. (2024). Unravelling the public spending-inclusive growth nexus: A PCA-GMM analysis in developing economies. ERU Research Journal,0(0), 1–17. https://doi.org/10.21608/erurj.2024.265595.1113 16 R. T. GUMEDE ET AL. Sarpong, B., & Nketiah-Amponsah, E. (2022). Financial inclusion and inclusive growth in sub-Saharan Africa. Cogent Economics & Finance,10(1), 2058734. https://doi.org/10.1080/23322039.2022.2058734 Scheffler, S. (2010). Equality and tradition: Questions of value in moral and political theory. Oxford University Press. Sen, A. (1985). Well-being and freedom. Journal of Philosophy,82(4), 185–203. https://doi.org/10.5840/ jphil198582467 Shrotryia, V. K., & Singh, S. V. P. (2020). Measuring progress beyond GDP: A theoretical perspective. Emerging Economy Studies,6(2), 143–165. https://doi.org/10.1177/2394901520983784 Sidrauski, M. (1967). Rational choice and patterns of growth in a monetary economy. The American Economic Review, 57(2), 534–544. Soegoto, H., Soegoto, S. W., & Meyer, D. F. (2022). The role of domestic investment, foreign investment and the number of micro small and medium-sized enterprises to reduce poverty in Indonesia. Journal of Eastern European and Central Asian Research (JEECAR),9(5), 901–913. https://doi.org/10.15549/jeecar.v9i5.1072 Stiglitz, J., Sen, A., & Fitoussi, J. (2019). Beyond GDP. Social Europe. United Nations. (2015). The millennium development goals report.https://www.un.org/millenniumgoals/2015_MDG_ Report/pdf/ MDG%202015%20rev Wiseman, J., & Brasher, K. (2008). Community wellbeing in an unwell world: Trends, challenges, and possibilities. Journal of Public Health Policy,29(3), 353–366. https://doi.org/10.1057/jphp.2008.16 World Development Indicators. (2023). https://databank.worldbank.org/source/world-development-indicators Worldwide Governance Indicators. (2023). https://www.govindicators.org/ Yolanda, Y. (2017). Analysis of factors affecting inflation and its impact on human development index and poverty in Indonesia. European Research Studies Journal,XX(4B), 38–56. https://doi.org/10.35808/ersj/873 Appendix Table A1. Selected countries. Upper middle-incomes countries High-income countries 1. Albania 1. Australia 20. Japan 2. Argentina 2. Austria 21. Latvia 3. Brazil 3. Belgium 22. Lithuania 4. Bulgaria 4. Canada 23. Luxembourg 5. China 5. Chile 24. Malta 6. Colombia 6. Croatia 25. Netherlands 7. Costa Rica 7. Cyprus 26. New Zealand 8. Dominican Republic 8. Czech Republic 27. Norway 9. Jamaica 9. Denmark 28. Poland 10. Mexico 10. Estonia 29. Portugal 11. Panama 11. Finland 30. Singapore 12. Paraguay 12. France 31. Slovak Republic 13. Peru 13. Germany 32. Slovenia 14. Romania 14. Greece 33. Spain 15. Russian Federation 15. Hungary 34. Sweden 16. South Africa 16. Iceland 35. Switzerland 17. Ireland 36. United Kingdom 18. Israel 37. United States 19. Italy 38. Uruguay Source: World Bank. COGENT ECONOMICS & FINANCE 17 Table A2. Data description and sources. Variables Description Source GDP In this case, GDP denotes gross domestic product (calculated at constant 2010 prices). GDP is calculated by incorporating the value of the final products and services produced in a specific country over time. WDI Education expenditure General government spending on education is expressed as a percentage of GDP, annually. IMF-GFS SOXC Health expenditure General government spending on health is expressed as a percentage of GDP. IMF-GFS SOCX Social protection spending General government spending on social protection is expressed as a percentage of GDP, annually. IMF_GFS SOXC Age dependency Reflects the ratio of dependents between 15 and 64 years old to individuals of the working age population (15-64 years old). WDI CO2 emission Presenting carbon dioxide emissions (kg per 2017 ppp $ of GDP) generated from fossil fuel combustion and cement production. WDI Employment Employment in services (% of total employment). This sector comprises several industries such as real estate, finance, insurance, wholesale and retail trade, restaurants, lodging, and hotels; in addition to transportation, storage, and communications. WDI Gross capital formation Gross capital formation (%GDP) represents the fixed assets of the economy. WDI Inflation An economy’s overall increase in the cost of goods and services is known as inflation. Using the consumer price index (annual %) we measure inflation. WDI Control of corruption This indicator portrays the perceptions of how public power is being exploited for private gain, capturing both small-scale and massive corruption. WGI Government effectiveness Assesses the public’s perception of the government’s commitments to delivering public services, its quality, and its independence from political pressures. WGI Rule of law Portrays the trust and compliance of people with the rule of law, particularly concerning contract enforcement, property rights, court enforcement, along crime and violence risks. WGI Political stability This indicator captures people’s views of the potential of political instability. WGI Life expectancy Life expectancy, at birth (years) Depicts life expectancy of infant at birth assuming that mortality rate recorded at birth remains constant. WDI Basic drinking water Access to improved drinking water (% of the population). WDI Sanitation Access to safety managed sanitation services (% population) WDI 18 R. T. GUMEDE ET AL.