Measuring and Fostering the Progress of Societies:
A New Approach for CIS and Eastern European Countries
Moscow, 29-30 September 2008
The Modified Human Development Index (MHDI) and its Applicability for Measuring Progress of Societies, the Hungarian Experience
Péter JÓZAN M.D, Ph.D., D.Sc.
Hungarian Central Statistical Office,
Center for Social Studies at the Hungarian Academy of Sciences
The OECD Global Project on Measuring the Progress of Societies builds on the first OECD World Forum on “Statistics, Knowledge and Policy”, held in Palermo in November 2004. Since then, a good deal of work has been done in research. The second OECD World Forum on “Statistics, Knowledge and Policy Measuring and Fostering the Progress of Societies” was held in Istanbul on 27–30 June 2007. On the second Forum, among others, the issues of policy making and democracy, political processes, accountability and civic engagement, people’s perceptions, subjective measures of well-being, challenges for national statistical offices and economic, social and environmental problems were discussed.
The Regional Conference on “Measuring and Fostering the Progress of Societies” to be held in Moscow on 29–30 September will discuss “A New Approach for CIS and Eastern European Countries”.
The projects of the Hungarian Central Statistical Office (HCSO), with respect to the Global OECD Project in general and to its New Regional Approach in particular are connected to the issue of Session 3. “What aspects of progress are the most important for CIS and Eastern European Countries… and how do we measure them?”
Here may I quote the economist-philosopher Amartya Sen[1]: “There is, in fact, something of a gap… between the tradition of evaluative statistics, on the one hand, and views of progress on the other, advanced by visionary social scientists.” He quotes Aristotle “who argued in Nicomachean Ethics that wealth is evidently not the good we are seeking; for it is merely useful and for the sake of something else.” According to William H. Draper[2] III, “Over the last three decades, a new concept has gained traction, positing that while growth is absolutely necessary for development, it is not sufficient… people must be at the center of development. The aim is to offer people more choices and opportunities to make their own decisions for long, healthy and creative lives.” The variables are, among others, “to have a sustainable livelihood and earn money, the ability to live a long and healthy life and to have access to decent schooling… Human development means more than the intrinsic value of personal fulfillment. An economy that hopes to stay competitive amid globalization must draw on everyone’s talents.”
Returning to Amartya Sen’s dichotomous approach, the Hungarian Central Statistical Office is attempting to put evidence based statistical data on a common denominator with the outlook of the “visionary social scientists”. This involves work in two areas to draw up a system of indicators measuring human development.
The outcome of the work is a well-structured set of nearly two hundred human development indicators appropriate – after suitable selection – for the study of any major social area. The system of indicators has already been subject to debate in the Hungarian Academy of Sciences, and on 18 September this year; the HCSO organised a one-day conference to discuss the main points of the working document, with the participation of Mr. E. Giovannini.
Work is also in progress on a new version of the UNDP human development index (HDI). The HDI is chiefly based on globalization considerations, and the modified version is designed to be more appropriate for measuring human development in industrial and post-industrial societies.
The OECD has an approach to measuring and fostering the progress of societies which is different to the UNDP’s measurement of human progress, but the two international organizations are in fact addressing similar challenges.
Below is an outline of the modifications made to the UNDP HDI to take account of European conditions, and a brief discussion of their applicability.
1. The index designed within the HCSO has the name of modified human developmentindex (MHDI).
2. The MHDI is a composite indicator with three components:
– Gross income (before taxation) (GI) per capita: GIPC;
– Percentage of adults in the population 25 years old and older with tertiary educational attainment (PATE)
– Life expectancy at birth.
3. Each component has equal weight in the composite index.
4. The GIPC was chosen in preference to GDP per capita (used by the UNDP) because it is available every year in local administrative units-1, Budapest districts and even the smallest villages (local administrative units-2), whereas GDP is calculated only for countries, regions and counties. At the level of regions (NUTS-2) and counties (NUTS-3), the correlation coefficients for the strength of relationship between per capita GDP and per capita GI are 0,96759, p<0,0001 and 0,92019, p<0,0001 respectively. This confirms the ability of the latter to substitute for the former. (Fig. 1.)
5.Calculation of the education index (EI) does not take account of the literacy of the adult population, because it may be assumed that the proportion of adults in the developed, and specifically European, countries who can read and write (adult literacy index) is nearly a hundred per cent. The qualifications of people who have passed through primary and secondary enrolment are also of limited significance[3]. The proportion of the population 25 years old and older with tertiary educational attainment, i.e. those with the highest level of skills, is an indicator which has been found capable of revealing differences in skills among observational units (including all local administrative units 1).
6. Life expectancy at birth expresses mortality in the measured period regardless of sex and age group. It is the most reliable, accurate, indirect, standardized indicator of health.
7. The MHDI is given on a conversion scale of 0 to 100 which combines the component indices representing three segments of human affairs.
8. The MHDI is best calculated for four-year periods, so as to ensure a sufficiency of data for the education indicator and life expectancy at birth in local administrative units-1 and the Budapest districts. The narrative interpreting the MHDI for each observational unit should be published every five years.
9. In the HCSO study, the MDHI was calculated for the years 2003-6. The per capita GI was taken for 2005, the proportion of the population 25 years old and older with tertiary educational attainment was taken from the 2001 census data, and the life expectancy at birth was calculated for the years 2003-6.
10. Level of development rankings may be set up for the regions (NUTS-2), the counties (NUTS-3), the local administrative units-1 (LAU-1) and the Budapest districts (see attached figures 2. 3.).
11. It is possible to calculate the strength of relationship between the MHDI, as the independent variable, and various dependent variables. These latter may be indicators of the political, economic, demographic, social, cultural, health and physical environment.
For example, strength of relationship may be quantified between the MHDI and indicators of
– party preferences;
– capital investment;
– unemployment;
– the ageing index;
– dependency ratio;
– fertility;
– internal migration balance;
– premature deaths (under the age of 70);
– deaths amenable by medical intervention (indirect indicator of health care effectiveness);
– cause-specific mortalities (cancer and cardiovascular mortality);
– morbidity;
– environmental pollution.
12. The MHDI may be used to track the long term development of a country or countries, regions, counties and local administrative units-1. Put another way, the well-being of nations may be quantitatively compared in specific years, and so time series may also be set up. Caution is required, however, in setting up time series. To obtain a reliable and accurate trend, especially in the longer term, it is necessary to take account of inflation (particularly if the rate exceeds the tolerable threshold), migration balance and the fact that percentages of local administrative unit-1 populations with tertiary attainment are only available in census years. The possibility of using the multi-year cumulative figure for matriculations in tertiary educational establishments instead of the percentage of people with tertiary attainment should be studied.
13. In 2006, the UNDP first published its attempts to find income inequalities within countries using the top and bottom deciles and the top and bottom quintiles with the Gini coefficient and by establishing the poverty rate (poverty being defined as income lower than 50 per cent of the national median income).
14. One component index of the MHDI, the per capita GI, examines income inequalities in an ecological context by distinguishing between local administrative units-1 in the top and bottom deciles and the top and bottom quintiles, and local administrative units-1 under the poverty threshold. The MHDI can be used to measure social inequalities in essentially the same approach as the HDI.
15. The MHDI is more appropriate to measure of well-being of nations in advanced, particularly European countries, than the index used by the UNDP, which incorporates aspects specific to less developed countries. Another advantage of the MHDI is that fewer data are required to calculate it than the HDI, and these data are available in every advanced country.
16. These features of the MHDI make it appropriate for international comparison of well-being of nations in a bilateral and multilateral context and among the countries of the European Union, the OECD and the United Nations Economic Commission for Europe. The educational index and the life expectancy at birth may be calculated with identical methodology in every observational unit in the advanced countries, yet in the case of per capita gross income a conversion to purchasing power parity (PPP) is necessary.
17. The relationships and also the inconsistencies between the MHDI and its component indices and the indicators of well-being and social inequalities are only really informative if interpreted in a widely understandable way. The quantified indicators and the narrative together provide the information which describes the real situation and thus contributes to rational decisions.
The MHDI has its limitations. It measures well-being, education and longevity with qualified objectivity, but says nothing about other aspects of human existence. The Human Development Report (HDR) gives information on this. It is hoped that the MHDI devised in the HCSO will contribute to production of the HDR.
Methodology:
Calculation of the modified human development index – MHDI – is based on the methodology of the United Nations Development Program (UNDP) human development index (HDI)
The formula for calculating the MHDI is:
Per capita gross income (before taxation), percentage of the population 25 years old and older with tertiary attainment and the life expectancy at birth are denoted , and respectively. The logarithm of income is taken. Converting to the variables , and , the human development index is given as a linear function of the variables . Designating , , , , and , the variables are subjected to the linear transformation
(i= 1, 2, 3).
The modified human development index is defined as the average of the values .
.
The MHDI was drawn up conceptually by Péter Józan and mathematically by László Radnóti.
Enclosed are maps and a simple regression diagram.
1
Fig. 1.
Relationship between GDP per capita and gross income (before taxation) GI per capita at NUTS-3 level (counties and Budapest), 2005
1
Fig. 2.
Modified Human Development Index (MHDI) in the Local Administrative Units1 (LAU1)
2003–06.
Fig. 3.
Modified Human Development Index (MHDI) in the districts of Budapest
2003–06.
1
[1], 2 The Measure of America, American Human Development Report 2008–2009, Foreword.
[2]
[3]Both adult literacy and primary and secondary enrolment are relevant indicators in less developed countries. There was therefore good reason for using them in the UNDP’s HDI. In developed countries, these indicators, especially adult literacy, are probably irrelevant.