Human capital, as the embodied knowledge and skills of citizens [1], is widely acknowledged as a key driver of a country’s economic growth [2, 3, 4, 5, 6, 7, 8, 9]. In recent years, the beyond GDP perspective posits that human capital determines sustainable human social welfare [10, 11, 12, 13]. In conjunction with produced (physical) and natural capital, human capital constitutes the wealth base for sustainable intergenerational development [14, 15, 16]. Notably, the U.N.'s 2015 Sustainable Development Goals (SDGs) identified three goals related to human development: education, health, and gender equality. These goals can only be achieved by effective human capital investments globally, which requires an in-depth retrospective investigation of human capital growth in the past decades.
The empirical accounting of human capital stock is complex. Schultz [17] stated that human capital is a resource that possesses both quantitative and qualitative dimensions. At the national level, Mincer [8] evaluated the quality of human capital through the formula of education attainment in the form of a discounted year of formal schooling. Subsequently, building on the later work of Jorgenson and Fraumeni [18, 19], Heckman and Klenow (1997), Klenow and Rogriguez-Clare [20, 21], Arrow & Dasgupta et al. [11] calculated human capital stock using a lifetime income-based approach. This theoretical formula illuminates three interlinked terms that affect human capital quantities and quality: education attainment, adult population size, and lifetime education return. This measurement framework has been applied in cross-country wealth accounting practices such as the UNU-IHP and UNEP [22] and Managi & Kumar's [15]Global Inclusive Wealth Tracing.
Researchers have become increasingly cognizant of the non-market benefits of investing in human capital [23, 24, 25, 26, 27]. However, accounting for human capital across countries to guide national investments in sustainable development is challenging. One of the primary challenges is assessing the quality of education. The Mean Years of Schooling (MYS) has traditionally been used as a proxy variable for assessing education levels. However, the MYS does not capture differences in the quality of education between countries [23, 28]. Detailed national information is required to reflect differences across countries accurately. Most fundamentally, MYS, as a backward measure, does not link current education investment with future returns when the lifetime income approach forwardly estimates the future value of human capital. In countries with market difficulties such as unstable education investment, capital substitution, and gender inequality, current education investment may not yield the same level of return in the future. Therefore, MYS-based measures of human capital may be overestimated.
The second challenge in assessing the noneconomic benefits of human capital is determining the quantitative attributes of human capital, such as population size and working lifetime. The population size is frequently viewed as an exogenous variable. However, demographic dynamics are intricately linked to the change in return of education [29]. Low-income countries with scarce human capital may be locked into a low human capital investment state while the population surges rapidly. Conversely, countries with substantial levels of human capital may experience aging and depopulation due to low fertility, which is closely correlated with the growth of returns on education. Additionally, the impact of migration and other demographic changes on human capital is a matter of concern [30, 31, 32]. Therefore, assessing human capital must consider the impact of demographic status.
Lastly, determining the value of returns on education and their fluctuations is crucial. The unit cost of human capital estimated based on compensation to workers may also be influenced by factors such as trade unions, taxes, and redistributive policy. Nevertheless, the level of return on educational investment should demonstrate the disparity in the quality of human capital across countries and regions. Moreover, the interrelation between education and health should also be valued [2, 33, 34, 17, 25, 35]. Several comprehensive human capital assessments have been developed to capture the non-monetary welfare benefits of human capital by incorporating education, demographic, and health information, such as the World Bank's Human Capital Index, IHME's Human Capital Index, and UNDP's Human Development Index. However, these indices need more insight into the link between investment in education and returns. It is essential to reconsider human capital measurement to address the aforementioned issues.
This paper presents an empirical framework that integrates education, demographic dynamics, health, and socioeconomic factors for cross-country human capital accounting. We propose measuring human capital using a lifetime income approach based on average life expectancy by stages. Firstly, we use Expected Years of Schooling (EYS) as a proxy variable to gauge educational attainment. This variable offers a forward assessment of educational level. We then calculate the total adult population as the stock of human capital based on the length of the EYS. We adopt the period equilibrium price as the comparable annual per unit cost for returns on education. We use Expected Years of Working (EYW) as a proxy to estimate lifetime income by education. The EYW illustrates average time losses due to unemployment and death of the adult population.
Our cross-country human capital accounting practice employs education information from UNESCO Institute for Statistics (UIS, 2020) surveys and national censuses. Demographic and lifetable data are sourced from the United Nations statistic division (UNSD, 2020). Labor participation data is obtained from the International Labor Organization (ILO, 2020), and GDP time series data is taken from UNSD (2021) to consider socioeconomic factors. Labor compensation data is obtained from various sources, including OECD, Eurostat, and Eora [36]. Through this methodology, we have estimated and tracked the human capital of 166 countries from 1990 to 2020.
Our estimates offer a comprehensive monetary human capital accounting and allow the evaluation of various perspectives on human capital progress, including those provided by educational investments, demography, health, and socioeconomic information. This accounting is beneficial for sustainability studies as it goes beyond just examining its relationship to economic growth.
In the subsequent sections of this paper, we first introduce the life expectancy-based lifetime income approach, the materials used, and the assessment framework. Then, in the third section, we present and explain the cross-country assessment results and discuss the implications. The final section provides the main conclusions and concludes the paper.