In practice, however, the measurement concentrates on a narrower definition. Knowledge, skills, and competencies certified by formal education have been the object of earlier research in the measurement of human capital. More recent developments have looked at other approaches to complement educational attainment indicators.
Measurement of human capital has implications for understanding the fundamental processes of societal development and economic growth. It also matters for estimating and understanding inequalities within societies. Measurements of human capital are important for the accountability of the education and health sectors. They help in accurately accounting for the costs and benefits of societal phenomena such as unemployment, and of the proposed policies to address those problems.
Human capital significantly determines a country’s consumption and production possibilities, today and in the future. People’s knowledge, skills, and competencies are “capital” in that they can be built up, but they can also decay, particularly during long periods of unemployment or sickness, or following shocks such as wars or migration. Failing to account for human capital could lead policy-makers to underweight investments in education, youth employment, and public health, with detrimental consequences for the future.
This is why in this chapter human capital is examined through the lens of sustainability. One of the motivations for focusing on human capital is a concern that, during and in the aftermath of the financial crisis of the late 2000s, estimates of the cost of the crisis did not reflect the decrease in human capital from high rates of youth unemployment, workers’ layoff, loss of firm-specific human capital, and lower spending on training by firms. If these costs were underestimated, the response to the crisis in terms of fiscal stimulus or investment in education and skills may have been too weak.
While there is widespread agreement about the importance of maintaining and increasing human capital to ensure sustainability, there continue to be discussions on the best way to measure it, on the advantages and disadvantages of different definitions of human capital (what should be included and what can be measured given data limitations), and on different methods to value it (see sidebar, “Approaches to Measuring Human Capital”).
APPROACHES TO MEASURING HUMAN CAPITAL
Traditionally, the most common approach to measuring aspects of human capital has been to use nonmonetary indicators of educational output, such as literacy or secondary school graduation rates. This type of indicator has the advantage of being widely available, both across countries and over time, even if data may not be fully comparable across countries. More recently, the indicators approach has been developed to take into account other aspects of human capital formation and stock characteristics. The use of standard classifications—for example, by type of education program—has improved the quality of these indicators.
Other approaches have aimed at producing a single summary measure of the stock of human capital in a country, expressed in monetary terms. These approaches include:
• The cost-based approach (Kendrick, 1976), where the stock of human capital is estimated as the depreciated value of the stream of past investments in human capital, such as teacher salaries and all other expenditures on education.
• The lifetime income approach (Jorgenson and Fraumeni, 1989, 1992a, 1992b), where the discounted value of the future labor income of individuals in the population for different education levels is calculated.
• The indirect or residual approach, which estimates the human capital stock as the difference between the discounted value of future consumption flows and the monetary value of other measured capital stocks. Because of its limitations, this approach is not recommended by the Guide on Measuring Human Capital (UNECE, 2016).
Summary measures of the stock of human capital may also be based on some combination of indicators and monetary measures. For example, the stock of human capital in a country may be measured as a weighted average of the mean years of schooling of different segments of the population (including those that are currently inactive), with weights based on estimates of the “rates of return to schooling” for various educational categories used to capture “quality.”
Progress Since the 2009 Stiglitz-Sen-Fitoussi Report
Some progress has been made in measuring human capital, as reflected in the Guide on Measuring Human Capital (UNECE, 2016). The guide provides reference and support for different strategies and approaches to measuring human capital, with an emphasis on preparing satellite accounts on human capital in line with SNA guidelines.
There is today general agreement in the statistical community on basic methodologies toward measuring human capital as related to education and labor market returns, though significant concerns remain. In general, in a national accounts context, the most appropriate measures are either cost approaches or the discounted lifetime income approach (in the tradition of Jorgenson and Fraumeni).4
While the lifetime income approach is appealing from a theoretical point of view, it requires detailed data and a number of assumptions; in particular, the value of human capital today depends on the assumptions you make about GDP growth in the future. This complicates the estimation, for example, of the impact of the economic crisis on future growth through the channel of reduced human capital. Other assumptions about the future must also be made, for example on life expectancy, and these assumptions can have large impacts on the overall estimate of the value of human capital.
The cost approach is based on past expenditure, but is also requires assumptions, in particular about depreciation of, and future returns to, human capital. However, for data availability reasons, the cost approach is most often preferred.
Several National Statistical Offices have recently undertaken initiatives to develop monetary measures of human capital, to be used alongside indicators of education quality and achievement. One common finding of these studies is that, whatever approach is used, the value of human capital is high compared to economic capital, even if the size of the discrepancies between estimates based on lifetime incomes and the cost approaches remains a puzzle (Liu, 2011). However, beyond the numerical estimates produced by these studies, considering educational expenditures as investment rather than consumption would have large impacts on how capital formation is defined and understood.
In addition, more recently, there has been substantial progress in the direct measurement of cognitive skills—in particular by the OECD through the PISA and, since 2011, the PIAAC survey (see sidebar below). The PISA survey, in particular, has played an important role by bringing human capital to the attention of policymakers in the educational community and beyond.
PIAAC AND PISA SURVEYS
Two OECD-sponsored instruments are increasingly used as a basis for computing human capital indicators:
The Programme for International Student Assessment (PISA) was run in 2006, 2009, 2012, and 2015. While all waves of PISA included tests in mathematics, science, and reading, the 2006 wave focused on science, the 2009 one on reading, and the 2012 one on mathematics. PISA testing also occurred in 2000 and 2003. In 2003 and 2012, tests were also offered in creative problem solving, while the 2012 wave included an optional test of financial literacy.
The Survey of Adult Skills, a product of the OECD Programme for the International Assessment of Adult Competencies (PIAAC), was designed to provide insights into the availability of some key skills in society and how they are used at work and at home. The first survey of its kind, it measures proficiency in several information-processing skills—namely literacy, numeracy, and problem solving in technology-rich environments.
Outstanding Issues and New Questions
Progress still needs to be made to improve the lifetime income approach for estimating the value of human capital and, more generally, to expand our understanding and measurement of human capital. Most of the analyses have thus far considered human capital as formal education or cognitive skills, and its returns as increased labor earnings. Future work ne
eds to expand the measurement of human capital to match the understanding that it is broader than education and cognitive skills, and that its returns are larger than individual earnings. The initial focus on education and labor returns was in part a function of data availability, and of the fact that this concept was more straightforward to operationalize when limited to these aspects. Even so, measuring human capital in these narrow terms suggests that human capital investments are undervalued.
It is now important to build on this foundation to go beyond cognitive skills, education, and remunerated activities. This broader perspective requires addressing difficult measurement questions such as how to measure noncognitive skills and nonmarket benefits, both individual and social, and understanding and measuring specific human capital and networks.
It also implies taking life expectancy and the demographic structure of the population into account, as those who live longer and healthier lives are more productive both in the market and in society. Migration also has to be included in the measurement of human capital: there is a cost for the sending countries when the better-educated population migrates.
Improving the Lifetime Income Approach
Quantifying the relationship of future productive potential to both levels of human capital and human capital investment is critical to establishing support for a more comprehensive treatment of human capital accounts, even though such a task faces clear methodological challenges. Better and more consistent estimates of the returns to education require, for example, longitudinal studies that can account for cohort effects (Boarini, Mira d’Ercole, and Liu, 2012).
In addition, measuring specific human capital is more complicated than measuring general human capital. Specific human capital includes, for example, firm-specific human capital, or networks, while general human capital includes schooling or nonspecific work experience. Failing to take on-the-job training into account may bias the estimate of the returns to formal schooling.
Understanding the gap between cost-based estimates and income-based estimates requires simultaneous estimation of the two. Implementing satellite accounts for human capital would help in better matching the two types of estimates.
However, many of the data needed for implementing the lifetime income approach are not available for some countries, and not necessarily harmonized. A better understanding of human capital, and improvements in its measurement, will come from more cross-country research. For example, countries vary in the structure of the earnings reported (e.g., time period considered, which particular criteria are included in the earnings definition) and in the reporting of educational attainment. Harmonized data, where available, would allow researchers to improve their understanding of the role played by education.
Finally, there is a need to better understand what the lifetime income approach can be used for, given the sensitivity of estimates to changes in assumptions about the future. While these estimates serve a very important role in demonstrating that human capital forms a very large component of wealth, and that spending on human capital should be considered as investment rather than consumption, it is less clear what practical use can be made of the approach in terms of planning and measuring sustainability.
HUMAN CAPITAL CONSEQUENCES OF RECESSIONS
A substantial concern with recessions is that unemployment, particularly youth unemployment, erodes human capital, or limits human capital acquisition through on-the-job training. If the full cost of recessions (the longterm lower GDP growth due to lower human capital) were recognized, policy responses might be stronger. While it is difficult to measure the loss of human capital due to recessions, these effects are likely to be important, as graduates who enter the labor market during a recession can be expected to have permanently lower incomes; these efforts are ignored by most applications of the lifetime income approach.
Expanding the Measurement of Human Capital and Its Returns
A strategy that values human capital by estimating the impact of education on lifetime income is also insufficient as it omits many important features on both sides of the equation: first, the human capital acquired outside of formal education, as well as noncognitive skills; and, second, the nonmarket benefits of human capital. A more comprehensive approach to human capital measurement would be important to lead policy-makers to recognize that education expenditure is a form of investment rather than consumption.
While the focus on formal education and market returns has been a function of pragmatically starting from where data availability and conceptual clarity were higher, it is important to increase data availability and conceptual clarity on human capital at all stages of capital formation, including its benefits and how it is embodied in individuals. For example, human capital investment takes place not only through education, but also through on-the-job training, parenting, and household production of nonmarket services (see sidebar, “Measuring and Valuing Unpaid Household Service Work”), as well as through participation in cultural activities. Destruction of human capital occurs, for example, in the presence of high youth unemployment, whose effects are not only lower consumption today but a lower long-run growth trajectory of the country tomorrow. Similarly, it is important to recognize that human capital is embodied in individuals not only through knowledge and cognitive skills, but also through noncognitive skills and traits. Its benefits encompass not only labor market returns, but higher subjective well-being, citizenship, caring, social trust, cooperation, and health. In this broader perspective, health care expenditures should be recognized as a kind of maintenance and repair flow for human capital, and better health conditions as nonmarket benefits from human capital. Human capital may also stimulate the accumulation of social capital: the norms and values that children develop at school will enable them to participate better in society as adults (OECD, 2010). In this broader perspective, developments in human capital can be seen in the context of the systems approach (outlined above in the section of that name), thanks to its links with developments in economic, social, and natural capital.
MEASURING AND VALUING UNPAID HOUSEHOLD SERVICE WORK
Even if unpaid household work contributes to preserve and improve human capital, most of it is excluded from the production boundary of the SNA. Various efforts have been undertaken in recent years to measure the amount and type of work carried out in the household and to estimate the monetary value of this work. Some countries have started to value these activities in a Household Satellite Account, which provides important information on the economy and society. Time-use surveys are an important tool to capture the amount of time spent by individuals to provide nonmarket services that benefit other household members or society more generally. Putting a value to household work is, however, not straightforward since the work is unpaid and because it often results in intangible services. A UNECE Task Force (UNECE, 2017) recently released a set of guidelines for valuing unpaid household services.
Beyond the Average
As with income or wealth, it is important to go beyond the average when examining human capital: there are important inequalities in human capital, which can vary by country and among population groups. In addition to understanding whether overall human capital is increasing or not, it is important to look at inequalities in human capital as these play an important role in shaping lifetime inequalities. For this reason, measures need to differentiate between adults and children, different groups in particular countries, and different household arrangements to understand how patterns are shifting over time.
Better measurement of these inequalities (in education and health care, for example) will contribute to a better understanding of inequality of opportunity. While gender inequalities in human capital are very important from a variety of perspectives (see sidebar), so are inequalities by income, race, caste, or religion.
GENDER INEQUALITIES IN HUMAN CAPITAL
While reducing inequalities in education would help reduce inequalities in life chances, gender-specific inequalities have
an even broader impact as they affect fertility decisions, the health of children, gender relations in the family with regard to power, and gender division of labor and authority within households.
The goal of achieving gender parity in education by 2005 has not yet been fulfilled, despite significant improvements. By 2011 only 60% of countries achieved this goal at the primary level, and 38% at the secondary level. In the world, more girls than boys are out of school: girls make up 54% of the total number of children out of school. In the Arab States, the share is 60%, unchanged since 2000 (UNESCO, 2015; UN Women, 2015). The gender parity index has increased dramatically in Southern Asia, where inequality was highest in 1990 and is now the lowest. However, while sub-Saharan Africa, Oceania, Western Asia, and North Africa have made progress, girls are still disadvantaged relative to boys regarding enrollment in primary education. Social inequality also widened, and this inequality often interacts with larger social and economic cleavages.
Attending school does not necessarily mean achieving basic literacy skills. It is a particular concern for poorer countries with insufficient teacher resources, but also in rich countries. The partial closing of the gender gap in primary education has contributed to reducing the incidence of illiteracy among women, but women still account for more than 60% of all illiterate persons in the world.
In secondary education, progress has been even more uneven across countries. On average, across the OECD, 43% of 25- to 64-year-olds have achieved an upper secondary or post-secondary, nontertiary degree. The improvement from the older to the younger cohorts is particularly large for women. Across the OECD in 2015, 37% of 55- to 64-year-old women, but only 15% of 25- to 34-year-old women, had no upper secondary degree (OECD, 2015).
For developing regions as a whole, the gender parity index for secondary education increased from 0.77 in 1990 to 0.96 in 2012. However, there are large differences between regions, with girls enjoying an advantage in Latin America and the Caribbean, but lagging significantly behind boys in sub-Saharan Africa, Southern Asia, Western Asia, and Oceania. Southern Asia stands out as the region where the greatest progress has been made, with the region’s index increasing from 0.59 to 0.93 between 1990 and 2012.
For Good Measure Page 41