The Human Development Index (hdi) Has Been Criticised For Various Reasons. What Are The Main Criticisms Of Hdi? How Would You Improve Hdi In Light Of These Criticisms?

The Human Development Index (HDI) is a measure of poverty and quality of life around the world. It is based on these kinds of measures that we understand, analyse and evaluate poverty and upon which we base most policy decisions, making what we choose to measure and how we choose to measure it critical in tackling this and other issues. For many years, these decisions were informed almost exclusively by some measure of income per capita, under the assumption that development relies heavily on material factors. In the 1990s and thanks to the work of Mahbub ul Haq and Amartya Sen, this material-centred approach was replaced by a more comprehensive view of what development entails, and thus, the alternative of the HDI was introduced. This measure combines life expectancy, education indicators and income per capita to produce a number between 0 and 1 that is meant to give a more accurate depiction of the level of development in a given country. This essay will begin by giving an overview of how the HDI is found and its basis on Sen s capabilities approach before exploring its main criticisms. It will then propose some solution to these and ultimately conclude that despite its flaws, the HDI is a marked improvement on previous methods of measuring development.

Firstly, it is worth considering the reasons why the HDI was introduced in the first place, namely, due the limitations of income-based measures of development. Income per capita varies widely (between around $750 and $124 410 per capita per year accounting for Purchasing Power Parity), implying that some countries are 165 times more developed than others, when in almost all other indicators, the differences in quality of life are not so broad.[1] Secondly, income measures can be inaccurate given incentives to underreport income and home-grown production is often omitted from the measure. Finally, although in some cases income might be a good indicator of development, it does not always align with what we intuitively associate with development: gender equality, medical infrastructure or good education services.

The HDI was therefore introduced to give a new perspective and ranking of countries that differs from their GNI per capita ranking. It was influenced by Amartya Sen s capabilities approach and his belief that a developed country is a free one, with freedom meaning the capability to lead the kinds of lives we have reason to value .[2] Although he believes that material factors such as income are important, they are only so insofar as they contribute to expanding our freedom and are not an end goal in themselves. He views political freedoms, economic facilities, social opportunities, transparency guarantees and protective security as the instruments through which freedom and development can be achieved.

The HDI does not provide quite so holistic a view of development, partly because of the difficulty in measuring many of the freedoms set out above but includes indicators beyond income per capita in order to capture the actual level of development in a country. It is constructed using three proxies to represent a long and healthy life, knowledge and a decent standard of living : life expectancy at birth, a mean of years of schooling and expected years of schooling and the log of GNI per capita adjusted at PPP.[3] The log of income is used to account for its diminishing marginal utility.

For each indicator, minimum and maximum goalposts are set, with the minimum values being subsistence values and the maximum values being the highest observed in the time series. The number assigned to each country is measured by (actual value minimum value) / (maximum value minimum value) for each indicator, and then the geometric mean of the indicators is found to give the final HDI number. In the last ten years, the HDI has become more comprehensive to include different variables of interest, e.g. gender inequality using different proxies and a discounted HDI using Atkinson s family of inequality measures. It can also be measured in groups within countries, across regions and reflects the variation between rural and urban areas. Countries are then grouped into low-HDI (0-0.549 Niger Syria), medium-HDI (0.55-0.699 Solomon Islands Marshall Islands), high-HDI (0.70-0.799 Egypt Serbia) and very high-HDI (0.80-1 Seychelles Norway).[4]

This essay will now turn to some criticisms of HDI and how these might be addressed. Firstly, the HDI along with other unidimensional methods of measurement is criticised by Alkire and Foster. Although it is a composite measure, the individual indicators rely on the validity of the aggregate variable in representing the actual resources of achievement of people and the actual trade-offs among component variables .[5] In reality however, the component variables of indicators might ostensibly be unrelated (or not related enough though this does not apply so much to the HDI) and be measured in different units.

The indicators represent different categories of needs that some might argue, cannot be merged in a significant way. In fact, this is another criticism of HDI. The way that it is computed implies that each indicator is a perfect substitute for another: i.e. an increase in income and a decrease in life expectancy could average out to give the same HDI number. Furthermore, the decision of the indicators themselves implies a value judgement. Perhaps there are other better measures of a long and healthy life than simply life expectancy, as the quality of those years could vary widely. If the knowledge indicator is itself composite, why is that so? And why is it not the case for the other two? The weights given to each factor are also somewhat arbitrary given their different units of measurement. It seems somewhat arbitrary to grant theme equal weights when income and years of schooling may have a different impact on development.

These criticisms have been prevalent from the inception of the HDI and are part of the reason that income-based measures are still very widely used thanks to their clarity and practicality. However, there were attempts to address them in the 2010 review of the HDI. First, to make the knowledge indicator more accurate, literacy rates were replaced by the new mean of the other indicators, as literacy is measured differently across countries. The change from a geometric mean to an arithmetic mean also reflected the idea that large disparities across the indicators would result in a lower HDI.

The choice of indicators might be criticised but in order to create such an index, it is the case that some value judgement has to be made and these indicators seem like a good enough choice given that some choice is inevitable. The addition of the gender equality by including fertility rates, or female share of the workforce would be an improvement and I believe should be included in the regular HDI. Similarly, in terms of the weights, unless one can find a very compelling reason why they should be weighed differently, then choosing the same weights might be pertinent. The criticisms regarding unidimensional methods are valid, but one of the HDI s strengths is its communicability and succinctness, and perhaps this must sacrifice the additional precision of multidimensional methods. Finally, it is worth considering that the HDI is somewhat constrained by data availability and lack thereof. Many signs of development such as the quality of education are unobservable in many instances, and so the HDI must use indicators that are accessible and universal. Under this constraint, perhaps the HDI cannot be improved very considerably beyond the changes that have already been made.

In conclusion, the Human Development Index was introduced as an alternative to measures that only included economic factors. Based on the works of Sen and ul Haq, it is an attempt to measure development not just in terms of how much money people have, but also how educated they are and how long they live. Its construction can be criticised in a number of ways, particularly in the choice of indicators and the way they are aggregated, but the HDI remains an understandable and more comprehensive perspective on development that can perhaps shed light on what policy decisions are more pertinent where. It also concedes that wealth is not always a measure of development (as exemplified by the case of Nauru). Ultimately, HDI requires a value judgement that one can disagree with, but I believe that this judgement is more accurate than many of its alternatives.

[1] https://databank.worldbank.org/data/download/GNIPC.pdf

[2] Sen, A., Development as Freedom, OUP, 1999, p285

[3] UNDP Human Development Report 2013 Technical Notes

[4] UNDP Human Development Report 2019

[5] Alkire and Foster, (2011), Understandings and misunderstandings of multidimensional poverty measurement , Journal of Economic Inequality