B: Technical Notes
This section provides further details of the methodology used in the construction of the Global Human Capital Index.
The Global Human Capital Index covers 21 unique indicators, which translate into 44 distinct data points once disaggregated by age group as appropriate in the Capacity and Deployment subindexes (Table B1). To be included in the Index an indicator must have available data for at least half (50%) of the sample countries. Values for each of the indicators come from publicly available data originally compiled by international organizations such as the International Labour Organization (ILO) and the United Nations Educational, Scientific and Cultural Organization (UNESCO). In addition to hard data, the Index uses qualitative survey data from the World Economic Forum’s Executive Opinion Survey. While an overview of the Index indicators is provided in Table B1, detailed descriptions, technical names and sources are included in the separate User’s Guide: Exploring the Global Human Capital Index Data section.
Table B1: Structure of the Global Human Capital Index
The Capacity subindex features four common measures of formal educational attainment. These capture the percentage of the population that has achieved at least primary, (lower) secondary or tertiary education, respectively, and the proportion of the population which has a basic level of literacy and numeracy. A workforce that is highly educated or at least has a solid foundational level of learning is much better prepared to adapt to new technologies, innovate and compete on a global level. Countries that have predominantly a primary level of education only are more likely to be constrained by low levels of income, fewer opportunities for future development for individuals, and, potentially, displacement of their workforce by new technology. Noticeably, many low-income countries have made remarkable strides in the past decades, with the result that the educational attainment of their younger age groups is frequently significantly higher than that of their older age groups, nearly drawing level with higher income countries in some cases.
Indicators and data sources
- Literacy and numeracy: Percentage of the population with the ability to both read and write and make simple arithmetic calculations. Source: UNESCO, Institute for Statistics, data from 2015 or latest available (accessed May 2017).
- Primary education attainment rate: Percentage of the population with at least a primary education (ISCED 1). This data is cumulative, which means that those with secondary education and above are counted in the primary education figures. Therefore, total figures across more than one category may add up to more than 100%. Source: Lutz et al., IIASA/VID Educational Attainment Model, GET Projection, 2015, Wittgenstein Centre for Demography and Global Human Capital (accessed May 2017); Barro and Lee, “A New Data Set of Educational Attainment in the World,” 1950-2010, Journal of Development Economics, 2010, http://www.barrolee.com (accessed May 2017).
- Secondary education attainment rate: Percentage of the population with at least a secondary education (ISCED 2–4). This data is cumulative, which means that those with tertiary education are counted in the secondary education figures. Therefore, total figures across more than one category may add up to more than 100%. Source: Lutz et al., 2015 and Barro and Lee, 2010, op. cit. (accessed May 2017).
- Tertiary education attainment rate: Percentage of the population with at least a tertiary education (ISCED 5–8). Lutz et al., 2015 and Barro and Lee, 2010, op. cit. (accessed May 2017).
The Deployment subindex measures how many people are able to participate actively in the workforce as well as how successfully particular segments of the population—women, youth and older people, those who tend to be particularly inefficiently engaged in labour markets—are able to contribute. Included in the Index—across all age groups except the under-15 age group—are the respective age group’s labour force participation rate, unemployment rate and underemployment rate. Including both those currently employed as well as people actively looking for work, a country’s labour force participation rate is the broadest measure of the share of its people participating in the labour market. Unemployment rates capture the subset of this group that is currently out of a job but would like to work. The underemployment rate is the share of those currently employed who would be willing and available to work more, thereby contributing their knowledge and experience more fully, and predominantly concerns people in involuntary part-time or fixed-term employment arrangements.
In addition to these three base measures, the Deployment subindex captures a key concept that is particularly critical for a specific segment of the population: a measure of the gender gap in economic participation, as this remains a critical weakness in most labour markets around the world. There is now widespread recognition of the individual and societal returns of increasing female labour force participation and employment rates for a strong and balanced economy. For countries with a shrinking working-age population, accelerating the integration of this well-educated and capable segment of the population is becoming ever more urgent.http://reports.weforum.org/global-gender-gap-report-20141
Indicators and data sources
- Labour force participation rate: Percentage of the population that engages actively in the labour market, either by working or looking for work. Source: ILOSTAT, Modelled Estimates, Labour force participation rate by sex and age, July 2017.
- Employment gender gap: Ratio of female labour force participation rate over male value, expressed as a percentage, capped at parity. A value equal to one indicates gender parity; a value less than one indicates a disparity in favour of men. Hence, the Index rewards countries that reach the point where outcomes for women equal those for men, but it neither rewards nor penalizes cases in which women are outperforming men. Source: ILOSTAT, Annual Indicators, Employment-to-population ratio by sex and age, data from 2014 or latest available (accessed May 2017).
- Unemployment rate: Number of unemployed persons as a percentage of the total number of persons in the labour force. The unemployment rate is a measure of the underutilization of the labour force. It reflects the inability of an economy to generate employment for those persons who want to work and are actively seeking work. It is thus an indicator of the efficiency and effectiveness of an economy to absorb its labour force and of the performance of the labour market. Source: ILOSTAT, Annual Indicators, Unemployment by sex and age, data from 2015 or latest available (accessed May 2016).
- Underemployment rate: Number of persons who, given the opportunity, are willing and available to work additional hours as a percentage of the total number of in employment. It includes persons wishing to take on another job in addition to their current employment, to replace their current employment with another one with increased hours of work, to increase the hours of work of their current job(s), or any combination of the above. The underemployment rate is a measure of the underutilization of the labour force. It signals employment perceived as inadequate (by the worker) and complements other indicators of labour underutilization such as the unemployment rate. Source: ILOSTAT, Annual Indicators, Time-related underemployment rate by sex and age, data from 2014 or latest available (accessed May 2017).
Social and economic marginalization still denies education to many. Access to education for today’s children and youth—the future workforce—is captured using net adjusted enrolment rates for primary school and net enrolment rates for secondary school, as well as through gross tertiary enrolment ratios and a measure of the education gender gap at the secondary enrolment level, for the under 15 and 15–24 age groups. The net enrolment ratios capture all children and youth who are enrolling at the appropriate age for that school level. As young adults in the 15–24 age group with completed secondary education face a choice between tertiary studies, acquiring further specialized vocational skills or entering the labour market, the Index includes a measure of enrolment in vocational training programmes, without making a value judgement between these three options in terms of index scoring.2
Although enrolment measures show exposure to learning, they don’t capture the quality of these learning environments and may be incomplete on their own.3
However, internationally standardized outcome measures of education quality—such as the OECD’s PISA test or the TIMMS and PIRLS tests—are available for a limited number of countries only. In the interest of broader country coverage, the Index therefore includes two qualitative indicators from the World Economic Forum’s Executive Opinion Survey on the quality of primary education and on how well the education system as a whole meets the needs of a competitive economy, as assessed by a country’s business community.
In addition, skills mismatches may arise when, irrespective of the level of qualifications individuals hold, fields of study do not match those demanded by employers. For example, employers in many countries point to shortages linked to too few young people studying science, technology, engineering or mathematics, and thus report skill shortages in specific professions. A broad base of skills is particularly important in ensuring a country’s resilience and adaptability in the face of the exponential technological and economic changes currently underway.4 The Index thus includes an assessment of the skill diversity of a country’s recent graduates as a proxy for the range of expertise available to a country.
The aspect of formal staff training is covered via survey response data from the World Economic Forum’s Executive Opinion Survey, which—as for the case of the education quality questions—should be treated as an indirect outcome measure of the extent and quality of such training received.
Indicators and data sources
- Primary education enrolment rate: Net adjusted enrolment rate, which refers to the percentage of children in the official primary school age range who are enrolled in either primary or secondary education. Source: UNESCO, Institute for Statistics, data from 2014 or latest available (accessed May 2017).
- Quality of primary schools: Response to the survey question, “How would you assess the quality of primary schools in your country? (1 = poor; 7 = excellent, among the best in the world)”. Source: World Economic Forum, Executive Opinion Survey, 2016–2017.
- Secondary education enrolment rate: Percentage of children in the official age range for lower secondary education who are enrolled in secondary education. In many education systems with compulsory education legislation, completion of lower secondary education coincides with the end of compulsory general education, intended to result in the full acquisition of basic skills. In most countries, the educational aim is to lay the foundation for lifelong learning. Source: UNESCO, Institute for Statistics, data from 2014 or latest available (accessed May 2017).
- Secondary enrolment gender gap: Ratio of female enrolment rate in lower secondary education over male value, expressed as a percentage, capped at parity. A value equal to one indicates gender parity; a value less than one indicates a disparity in favour of boys. Hence, the Index rewards countries that reach the point where outcomes for women equal those for men, but it neither rewards nor penalizes cases in which women are outperforming men. Source: UNESCO, Institute for Statistics, data from 2014 or latest available (accessed May 2017).
- Vocational education enrolment rate: Technical/vocational enrolment as a percentage of total enrolment in upper secondary education (ISCED 3), following completion of compulsory general (basic) education. In many countries, programmes at the upper secondary education level are more specialised and offer students choices and diverse pathways. Source: UNESCO, Institute for Statistics, data from 2014 or latest available (accessed May 2017).
- Tertiary education enrolment rate: Total enrolment in tertiary education (ISCED 5–8), regardless of age, expressed as a percentage of the total population of the most recent five-year age cohort that has left secondary school. Tertiary education builds on secondary education, providing learning activities in specialized fields of study. It aims at learning at a high level of complexity and specialization. Tertiary education includes what is commonly understood as academic education but also includes advanced vocational or professional education. Source: UNESCO, Institute for Statistics, data from 2014 or latest available (accessed May 2017).
- Skill diversity of graduates: Measure of the diversity of fields of study of recent tertiary education (ISCED 5–8) graduates in a country. Calculated as a Herfindahl-Hirschman Index (HHI) of concentration among the broad fields of study recognized by UNESCO’s International Standard Classification of Education (ISCED 2011). A perfectly equal distribution of graduates among disciplines would result in a normalized HHI value of 0.090, while a complete concentration of graduates in just one discipline would result in an HHI value of one. Source: World Economic Forum calculation; using data from UNESCO, Institute for Statistics, data from 2014 or latest available (accessed May 2017).
- Quality of education system: Response to the survey question, “How well does the educational system in your country meet the needs of a competitive economy? (1 = not well at all, 7 = very well)”. Source: World Economic Forum, Executive Opinion Survey, 2016–2017.
- Extent of staff training: Response to the survey question, “To what extent do companies in your country invest in training and employee development? (1 = hardly at all, 7 = to a great extent)”. Source: World Economic Forum, Executive Opinion Survey, 2016–2017.
Know-how concerns the extent of human capital acquisition in the workplace through learning-by-doing, tacit knowledge, exchange with colleagues as well as through formal on-the-job learning, continued education and staff training. Economic complexity is a measure of the degree of sophistication of a country’s “productive knowledge” as can be empirically observed in the quality of its export products.5 Given that age-disaggregated measures of this concept were not available, the decision was made to locate the corresponding indicators within the 25–64 age group, which encompasses the bulk of the working population but does not imply that these processes are not similarly important for the other age groups. In addition, the Index measures the current level availability of high and mid skilled opportunities and, in parallel, employer’s perceptions of the ease or difficulty of filling vacancies.
Indicators and data sources
- High-skilled employment share: Number of persons, both sexes, employed in occupations with tertiary (ISCED 5–8) education requirements as a percentage of the total number of employed persons. Source: International Labour Organization, Trends Econometric Models, October 2014 (accessed May 2017).
- Medium-skilled employment share: Number of persons, both sexes, employed in occupations with at least secondary (ISCED 2–4) education requirements as a percentage of the total number of employed persons. This data is cumulative, which means that persons employed in occupations with tertiary (ISCED 5–8) education requirements are also counted in the medium-skilled employment figures. Source: International Labour Organization, Trends Econometric Models, October 2014 (accessed May 2017).
The Index’s methodology follows that of the ILO, which has aligned each of the major occupational groupings of the International Standard Classification of Occupations (ISCO-08) to one of four skill levels, ‘defined as a function of the complexity and range of tasks and duties to be performed in an occupation’.6 Each skill level has, in turn, been aligned to the level of formal education of the International Standard Classification of Education (ISCED-97) typically required for competent performance in the occupation, resulting in the following classification scheme:
Skills levels 3 & 4 = Tertiary education requirements (ISCED levels 5–6)
Managers, professionals and technicians
Skills level 2 = Secondary education requirements (ISCED levels 2–4)
Clerical, service and sales workers
Skilled agricultural and trades workers
Plant and machine operators and assemblers
Skills level 1 = Primary education requirements (ISCED level 1)
- Ease of finding skilled employees: Response to the survey question, “In your country, how easy is it for companies to find employees with the required skills for their business needs? (1 = extremely difficult, 7 = extremely easy)”. Source: World Economic Forum, Executive Opinion Survey, 2016–2017.
- Economic complexity: Measure of the breadth and depth of productive knowledge and skills of a country’s workforce, as embodied in the complexity of its export products. Derived from the Atlas of Economic Complexity, which aims to capture the extent to which ‘modern societies can amass large amounts of productive knowledge because they distribute bits and pieces of it among its many members. … Thus, individual specialization begets diversity at the national and global level. Our most prosperous modern societies are wiser, not because their citizens are individually brilliant, but because these societies hold a diversity of know-how and because they are able to recombine it to create a larger variety of smarter and better products’. Source: Hausmann, R., Hidalgo, C., et al., The Atlas of Economic Complexity, http://atlas.cid.harvard.edu/rankings, data from 2014 (accessed May 2017).
For the majority of indicators, a reference point/interval-based scale has been used to convert the values of the raw data into a common metric. Each indicator is assigned a logical minimum and maximum value and all raw data points are then expressed as the gap towards attainment of the ideal value, on a scale from 0 to 100. As many of the concepts measured by the Global Human Capital Index are expressed as percentage rates for the corresponding age group, their “distance to the ideal” can be clearly defined and takes on intuitive minimum and maximum values. For example, the Primary enrolment rate indicator has a logical maximum value of 100% and a higher score reflects a more desirable situation.
Most indicators range between 0 and 100, but overall three additional interval types exist. One, survey responses are on a 1 (worst score) to 7 (best score) scale, which is applicable to the Quality of primary schools, Quality of education system, Staff training, and Ease of finding skilled employees indicators. Two, on the indicator that captures skill diversity, data is normalized on a 0.090 (best score) to 1.000 (worst score) scale. Three, on the indicator that captures economic complexity, data is normalized on a –3 (worst score) to 3 (best score) scale.
In some cases, the logical minimum value is numerically less than the logical maximum value, in others the logic is reversed. For skills diversity 0.09 is more than 1 and for under and unemployment 100 is logically less than 0.
These data points are converted to their standardized score based on the following formula:
There are two sets of indicators whose values are heavily skewed in one end of the scale—unemployment and education. They do not change over time in the same way and do not respond similarly to policy intervention. While infrastructure and access can drive education results across the scale, sometimes from very low attainment to high attainment, unemployment is more skewed. We therefore apply a logarithmic transformation to one but not the other using a natural logarithm.7 Figure B1 highlights the transformation of the data this formula effects.
Across all indicators, the final scores can be roughly interpreted as a percentage, reflecting the degree to which human capital has been optimized in a given country.8
There are a number of limitations to this approach to standardization. The logical minimum and maximum values assigned to each indicator are independent of the spread of the range of indicator values, so an indicator that has a higher value range will have a greater impact on the country’s overall Index score relative to an indicator that has a lower value range. This can be exacerbated with missing data. While recognizing this limitation, the approach of standardizing against a reference was found to be the most technically sound given the Index’s choice of indicators and overall purpose, particularly as it enables countries’ progress to be tracked year on year, independently as well as relative to the performance of other countries.9
Figure B1: Logarithmic transformation of scores for the Unemployment rate indicator
The four thematic dimensions serve as the subindexes and are weighted equally while the age-group specific data within these is weighted by population. We first aggregate each age group within the subindex and then derive the score for the subindex by weighting each bundle by the specific distribution of the country’s population. As a consequence, the index is now more nuanced in highlighting strengths and weaknesses in capacity and deployment by taking different demographic structures into account.
Missing data and country coverage
To enable valid comparability across countries, we have established the following exclusion barriers for data points:
- For each age group within the Capacity subindex, at least two out of four indicators.
- For each age group within the Deployment subindex at least two out of four indicators.
- In the Deployment subindex at least six out of nine indicators.
- In the Know-how subindex at least three out of four indicators.
We have excluded selected countries due to concerns about recent data quality and excluded all countries in which the Executive Opinion Survey has not been conducted during the past year, which often includes those in which civil or military unrest does not allow accurate or relevant measurement.
Data older than 10 years was considered to be of insufficient relevance for the Index.10 In general, the Global Human Capital Index does not impute missing data. A few exceptions were undertaken in order to enable countries to meet the minimum coverage criteria for inclusion in the Index after reviewing sensitivity analysis.
The literacy rate of those in OECD countries aged 15–24 was assumed to be 100 or not distinguishable from 100. We did not make the same assumption for older age groups, who have traditionally had less access to education. Across a number of indicators we established a preferred and secondary data source, for example, in educational enrolment using net and gross values and in deployment data switching between estimates and projections and yearly indicators. Two instances of old data were assumed to have remained unchanged (Nigeria’s vocational education enrolment rate and the UAE’s tertiary enrolment rate).
The 2017 edition of the Index covers 130 countries. The terms “country”, “economy” and “nation” as used in the Global Human Capital Report do not in all cases refer to a territorial entity that is a state as understood by international law and practice. The term covers well-defined, geographically self-contained economic areas that may not be states but for which statistical data are maintained on a separate and independent basis.
Comparison to the 2016 edition
Since the release of the first edition of the Index in 2013, much thoughtful feedback has been received.11 In addition, we have continuously monitored data sources and methodological updates in the wider human capital literature for opportunities to further refine the Index. As a result the latest edition of the Index incorporates some notable changes that are aimed at streamlining key concepts and enhancing the reader’s comprehension of the dynamics driving the growth of human capital. These are described below.
As in previous editions, the 2017 Report groups results across five age groups (0–14, 15–24, 25–54, 55–64, 65+). While retaining the Report’s traditional focus on maximizing human capital across the age range, indicators have been reorganized into four distinctive thematic subindexes: Capacity, Deployment, Development and Know-how. By contrast, previous editions had grouped indicators into only two themes: Learning and Employment. Under this year’s enhanced framework, the indicators “literacy and numeracy” and “employment gender gap” have also been distributed across the age range, expanding the comparability between age groups within the Capacity and Deployment subindex.
In the 2016 edition of the Index, the age groups acted as de-facto subindexes and the Index was derived by weighting each age group by the distribution of the global population. In this enhanced 2017 edition, the four thematic dimensions are the subindexes and are weighted equally while the age groups are population-weighted dimensions within the new subindexes. Therefore, to calculate the Index, we first aggregate each age group within the subindex and then derive the subindex score by weighting each bundle by the specific distribution of the country’s population. As a consequence, the Index is now more nuanced in highlighting strengths and weaknesses in talent capacity and deployment by taking different demographic structures into account.
The indicators Long term unemployment rate, Child labour and Healthy life expectancy have been omitted from the core Index in this year’s edition for conceptual focus and to prevent overlaps with other indicators in the Index. They remain important contextual factors and covered within the main chapter. In addition, two indicators have been removed from the index due to consistently weak and/or irregular data coverage: Basic education survival rate and Over- and Under-education. Finally, contextual data in the Country Profiles have been updated to include information on wages, productivity and social security.
Acemoglu, D. and D. Autor, “What Does Human Capital Do? A Review of Goldin and Katz’s The Race Between Education and Technology”, NBER Working Paper 17820, The National Bureau of Economic Research, 2012, http://www.nber.org/papers/w17820.
Angrist, N., H. Patrinos and M. Schlotter, “An Expansion of a Global Data Set on Educational Quality: A Focus on Achievement in Developing Countries”, World Bank Policy Research Working Paper 6536, World Bank, 2013.
Barro, R. and J.W. Lee, “A New Dataset of Educational Attainment in the World, 1950–2010”, NBER Working Paper 15902, The National Bureau of Economic Research, 2013, http://www.nber.org/papers/ w15902.
Boarini, R., M. Mira d’Ercole and G. Liu, “Approaches to Measuring the Stock of Human Capital: A Review of Country Practices”, OECD Statistics Working Papers, 2012/04, Organisation for Economic Co-operative Development, 2012, http://dx.doi.org/10.1787/5k8zlm5bc3ns-en.
Brynjolfsson, E. and A. McAfee, The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies, 2014, http://www.secondmachineage.com.
Cuaresma, J.C. and T. Mishra, “The role of age-structured education data for economic growth forecasts”, Journal of Forecasting, vol. 30, issue 2, 2011, pp. 249–267.
Delgado, M., D. Henderson and C. Parmeter, “Does Education Matter for Economic Growth?”, IZA Discussion Paper No. 7089, Institute for the Study of Labor, 2012, http://ftp.iza.org/dp7089.pdf.
Fengler, W., “Don’t let perfect be the enemy of good: To leverage the data revolution we must accept imperfection”, Brookings, 14 April 2016, http://www.brookings.edu/blogs/future-development/posts/2016/04/14-big-data-revolution-technologies-fengler.
Hausmann, R., C. Hidalgo, et al., “The Atlas of Economic Complexity”, Centre for Economic Development at Harvard University, http://atlas.cid.harvard.edu.
International Labor Organiztion (ILO), International Standard Classification of Occupations Structure, group definitions and correspondence tables, ISCO-08, Volume 1, 2012.
Lutz, W., et al., “Validation of the Wittgenstein Centre Back-projections for Populations by Age, Sex, and Six Levels of Education from 2010 to 1970”, IIASA Interim Report IR-15-008, International Institute for Applied Systems Analysis, 2015, http://www.iiasa.ac.at/publication/more_IR-15-008.php.
Nardo, Michela, Michaela Saisana, Andrea Saltelli and Stefano Tarantola, “Tools for Composite Indicators Building”, European Commission Joint Research Centre (JRC), 2005, http://publications.jrc.ec.europa.eu/repository/bitstream/JRC31473/EUR%2021682%20EN.pdf.
OECD and European Commission (JRC), “Normalisation”, in: Handbook on Constructing Composite Indicators: Methodology and User Guide, 2008, http://www.oecd.org/std/42495745.pdf.
Omolewa, M., “Traditional African Modes of Education: Their relevance in the modern world”, International Review of Education, vol. 53, no. 5, 2007, pp. 593-612.
Reimsbach-Kounatze, C., “The Proliferation of ‘Big Data’ and Implications for Official Statistics and Statistical Agencies: A Preliminary Analysis”, OECD Digital Economy Papers No. 245, OECD, 2015.
Rizzo, M., Kilama, B. and M. Wuyts, “The Invisibility of Wage Employment in Statistics on the Informal Economy in Africa: Causes and Consequences”, The Journal of Development Studies, vol. 51, no. 2, pp. 149–161, 2015.
UNCTAD, Global Assessment of Sex-Disaggregated ICT Employment Statistics: Data Availability and Challenges on Measurement and Compilation, UNCTAD Technical Notes on ICT for Development no. 4, 2016.
UNESCO, International Standard Classification of Education (ISCED-2011), 2011, http://uis.unesco.org/sites/default/files/documents/international-standard-classification-of-education-isced-2011-en.pdf.
United Nations Population Division, World Population Prospects, 2017 Revision, 2017.
UNU-IHDP and UNEP (2014), “Part II: Human Capital”, in: Inclusive Wealth Report 2014. Measuring Progress Toward Sustainability, Cambridge University Press, 2014, http://mgiep.unesco.org/wp-content/uploads/2014/12/IWR2014-WEB.pdf.
Van Roy, V. and F. Rossetti, “JRC Statistical Audit on the 2013 Human Capital Index”, European Commission Joint Research Centre (Ispra, Italy) – Econometrics and Applied Statistics Unit, 2013.
Wright, J., “Making a Key Distinction: Real-Time LMI & Traditional Labor Market Data”, Emsi Blog, 7 February 2012, http://www.economicmodeling.com/2012/02/07/making-a-key-distinction-realtime-lmi-traditional-labor-market-data.