RESUMO
PIP: This discussion focuses on the advantages and disadvantages of constructing a summary measure of the multidimensional and dynamic concept of development. 2 indexes are referred to: the UN Development Programme; human development index (HDI) and the Physical Quality of Life Index (PQLI). PQLI weights equally life expectancy at 1 year, infant mortality, and literacy, regardless of the correlation of variables and the arbitrariness of the weights. PQLI also is assumed to measure the quantity of life, not the quality. HDI measures quantity and quality and includes life expectancy, literacy, and real GDP/capita, and may include a measure of human freedom. Objectivity is a major problem with any index. Assignment of weights is an example of arbitrariness without justification and the index is sensitive the weights assigned. There is a paradox between defending weights on the 1 hand and, if robustness of the index is assumed and components are correlated, then any single component could suffice. A more serious criticism of the HDI is the weighting of each rank order of the country by 1/3 and summing the weighted ranking of the 3 indicators. The flaw here is the problem of application of ratio scales on ordinal magnitudes. The rank correlation coefficient between real GDP/capita and life expectancy, real GDP/capita and literacy; and literacy and life expectancy are .90, .80, and .89, respectively. HDI is also correlated with GDP/capita (.87). Composite indexes are not sensitive to variability in components or the imbalance in components, e.g., a country with low means but high literacy. The goal should continue to be to develop a conceptually and methodologically acceptable summary measure. It is suggested that a necessary component is economic development. Countries may be ranked according to their level of economic development in order to measure their achievements in human development. A weighted distribution of income/capita is a better indicator of the economic well being of a country because it reflects the maldistribution of income. Access to available government goods and services by income groups should also be taken into account. Purchasing power parity should replace income/capita in nominal prices for international comparisons due to the presence of nontradable goods and services and distortions from exchange rate anomalies, tariffs, and taxes. Thus, a replacement for HDI is purchasing power parity adjusted real income/capita weighted by the distribution of income which uses the Gini Coefficients.^ieng
