Hacking the Human Development Index: Why UNDP's Latest HDI Scores For India Are Misleading

The Human Development Index (HDI) is a frequently cited metric of general socio-economic development. But what exactly is it, and how is it quantified and graded ? Where is India on this index ? What are the factors that affect the position on the list ? What are the most effective ways to improve a country’s HDI and ranking ? This article will attempt to dissect this metric and India’s performance.

The Human Development Index was introduced in 1990 by the United Nations Development Program (UNDP) as part of its annual report on socio-economic development, as a composite statistical measure that enabled countries to be measured on both an absolute and relative scale with respect to one another. It depends on a parametric representation of three key measures:

● Health - represented by life expectancy at birth

● Education - as represented by a measure of schooling and literacy

While this may be a very simplified representation of human development - and there are indeed critiques of the metric - the HDI score remains a broadly cited parameter of national development. An analysis of its deficiencies has led to several similar or complementary metrics such as the Inequality Adjusted HDI and Augmented HDI, as well as other indices such as the Gender Development Index . However, the standard HDI metrix remains by far the most significantly cited one, and this article will solely focus upon it.

Decomposing the HDI

The Human Development Index is a composite of three feature-scaled indices, that are used to subdivide performance into four category tiers - very high, high, medium and low human development. The tiers are expressed using the following cutoff scores for each sub-index as well as the overall index:

India currently has a reported 2023 HDI of 0.644 as reported by the United Nations Development Program (UNDP) in its 2023-24 report. This means that it is solidly in the middle of the Medium tier. In comparison, China ranks at 0.788. What leads to this score ? How can India best improve?

Life Expectancy Index

This index is a scaled expression of life expectancy at birth of a country, as a fraction of the age range 20 to 85 years. In other words, a life expectancy of 20 years or below means the index is 0.000 and 85 years or more means 1.000. It is thus expressed as:

For example, a life expectancy of 70 years translates to a LEI of 0.769.

We can thus estimate the life expectancy range that puts the LEI into various performance tiers:

Education Index

The education index is an unweighted mean of two parameters - mean years of schooling, and expected years of schooling.

Mean Years of Schooling Index

Mean years of schooling is an aggregate measure of the average years of education received by a country’s adult population. This measure is generated from historical enrollment data of a country’s adult population.

It is expressed as a scaled index that saturates at 15 years of education.

Thus a country with 0 years of average education for its adult population scores 0.000 and it scores 1.000 if it achieves 15 or more years of schooling for its adult population. Thus an MYS of 10 years of schooling for the adult population means a MYSI of 0.667 .

The MYS data can be imperfect - it doesn’t keep track of repeated years, or the relative quality of education. These are, of course, among the several critiques of a measure like HDI. However this article will focus on an abstract understanding of the index.

The tier cutoffs corresponding to MYS are therefore:

Expected Years of Schooling Index

The crucial difference with this index is that it estimates how many years of schooling current school age children will gain. It is measured from current enrollment data in primary, secondary and tertiary education tiers relative to the total population of each of those age groups.

It is expressed as a scaled index that saturates at 18 years of education. In other words, the index defines that current school going children should receive more education than the existing adult population whose education is defined by the MYSI figure.

Thus an EYS of 14 years of schooling for the current school going population means an EYSI of 0.778.

The tier cutoffs corresponding to EYS are therefore:

The Education Index is the unweighted mean of these indices:

Education Index (EI) =(MYSI + EYSI) / 2

Income Index

Income Index (II) =(log(minimum(75000,GNIPPPperCapita)) - log(100)) / (log(75000) - log(100))

A GNI PPP per capita of $15000 therefore translates to an II of 0.757.

The tier cutoffs corresponding to GNI PPP per capita are therefore:

Human Development Index

The actual human development index is the geometric mean of the LEI, EI and II figures. Thus,

HDI = GeoMean (LEI,EI,II)

India and the HDI

As mentioned earlier, India has a reported HDI of 0.644. This is obtained from the UNDP HDI report of 2023-24. This puts India squarely in the midpoint of the Medium category of HDI, quite far from the two higher tiers - High (0.700 to 0.799) and Very High (0.800-0.899).

Naturally this is not a very good figure. But what is the source of this figure? How contemporary is the data? What does India need to do to improve upon it? Let us deconstruct how the subindices came about.

India and the Life Expectancy Index of HDI

The UNDP 2023-24 report uses a life expectancy figure at birth of 67.7 years (Table 1, pg 276). The nearest data reference for this figure is WHO data that reports the trough of the COVID pandemic at 67.3. This in turn generates a Life Expectancy Index for India of 0.734, which classifies as High.  

However, this data is inconsistent with data and estimates reported by the Ministry of Health and Family Welfare. The MoHFW Annual Report 2022-23 reports data using trailing five year records. The latest figure is for 2020 using 2016-2020 data, which it reports as 70.0. In the MoHFW Central Bureau of Health Intelligence Report (July 2024) which reports a 2021-25 life expectancy figure of 69.4 for men and 72.7 for women, which - given the near par sex ratio today - translates to a combined life expectancy of 71.1. This translates to a LEI of 0.786 - close to the Very High tier which begins at 0.800.

As of 2025, various sources indicate that Indian life expectancy is approx 72-73 years, The midpoint of that translates to a LEI of 0.808, which puts India in the Very High category on the LEI as of the present day.

In summary, India already ranks at the midpoint of the High tier of LEI with what looks like clearly very pessimistic life expectancy data used in the UNDP report. A present day estimate generates an LEI at the threshold of the uppermost Very High tier. In effect, further progress on this metric will be consolidation within the Very High tier.

India and the Education Index of HDI

In the UNDP report, India has a Mean Years of Schooling of just 6.6 years, and an Expected Years of Schooling of 12.6 years. Applying the index calculations described earlier, this yields:

Both the mean and expected years of schooling figures are synthetic constructs obtained from a combination of time series data on gross enrollment, and the population within each age group. UNESCO publishes a methodology guide for computing this index. The MoHFW reports only the gross enrollment data, such as in its Annual Report 2022-23 (pg 64, table 2.1.2).

India and the Mean Years of Schooling Index

The mean years of schooling figure is broadly a problem for most developing and even (very) developed countries - it reflects past history in educational attainment over at least half a century, and due to this fact, it is the hardest metric to fix. Here are a collection of countries that have significant HDI figures but very low MYSI in the UNDP 2023-24 report:

Thus, it is possible to have High or Very High HDI attainment while still being years away from having a MYSI that is better than Low - as seen in the case of China or Turkey. This is because the MYSI is backward looking and penalizes a history of weak performance.

Let us assume a progressive improvement of the mean years of schooling in India to a conservative 7.2 years by 2025. This generates the following:

India and the Expected Years of Schooling Index

The EYSI parameter differs from the MYSI parameter primarily in being a forward looking estimate as opposed to a backward looking one. Further, a strong EYSI performance over time mitigates a country’s MYSI figure, as progressive generations report substantially higher education attainment than their parents.

Assuming that the 2025 expected years of schooling figure is 13.75, this yields an EYSI of 0.750 - which is stil in the High tier.

This estimated Education Index for 2025 for India is within the same ballpark as the UNDP 2023-24 report for China (0.644). In effect, India is doing quite well in this measure and the underlying requirement for progress is to maximize the EYSI and then to maintain it so that the impact shows on the MYSI.

India and the Income Index of HDI

However, as it turns out, this number is also conveniently picked up from the bottom of the COVID-19 pandemic. The period since then has been characterized by a growth in Indian per capita PPP GDP of 11.3%, by a distance the highest of any major economy in the world. India’s per capita income in PPP terms in 2025 is estimated to be approximately $12,000, up from $11,200 in 2024.

In other words, India is already established in the High tier of the Income Index, having rapidly traversed the range from middle of the Medium tier to the second quarter of the High tier within the past four years due to exceptionally high growth in purchasing power.

A caveat here is that the estimate uses reported PPP GDP per capita and not PPP GNI per capita data, since the former is more widely available. However, these two metrics are not very different, and GNI includes foreign income earned by residents that GDP does not account for.

Having now understood what the HDI is, its subcomponents and how India measures on these parameters, let us now put together a picture of how Indian HDI looks like today:

Looking at the above table, we see that India has transitioned tiers on every subcategory. The Life Expectancy Index has increased from High (0.734) to Very High (0.808) . The Education Index has risen from Low (0.569) to Medium (0.622) . The Income Index has gone from Medium (0.641) to High (0.723). The overall HDI is now over the threshold of 0.700 and now India is in the High HDI category.

This puts India in a peer group with several nations that are also in the same range of PPP per capita incomes - Indonesia, Philippines, South Africa, Vietnam, Paraguay - all of which have PPP GDP per capita figures in the $12,000-$18,000 range.

This also means that India is not fundamentally underperforming in health or education metrics relative to these countries, and that the progressive growth in income offers a proxy for growth in HDI as well. This is reflected by the correlation between the GNI PPP per capita rank and the HDI rank - India has a figure of -6 here in the UNDP 2023-24 report, which means that our income rank is very close to our HDI rank. In effect India is raising its HDI nearly as fast as it is raising its per capita income - which it has been doing extremely fast in the 2020s.

The delta in each sub-index also demonstrates that the most gain is to be had in the Income Index. For example, raising the life expectancy to 75 years will increase the LEI to 0.846, a 4.7% gain from the 0.808 figure for 72.5 years today. On the other hand, increasing per capita income to $18,000 will raise the Income Index to 0.784, an 8.4% gain .These are not unconnected - rapid income growth yields better health and education outcomes, thereby raising those indices more.

A State Level View of HDI

Data from 2022 shows that Indian state level HDI ranges from 0.758 (High) for Kerala - on par with Turkey - to 0.540 (Low) for Bihar, on par with Pakistan. But this high level view conceals a lot of detail.

In terms of life expectancy, every state in India has achieved or is close to High-tier status, with some reaching the Very High tier. This is plain by looking at 2021-25 estimates in the MoHFW Central Bureau of Health Intelligence Report (page 36, table 1.2.5(b)) from mid 2024.

A decade ago, only one state of India had a mean years of schooling exceeding 10.5 years that characterizes High tier attainment - not Kerala but Manipur. While the author has not been able to obtain more granular data for the present day, gross enrollment data has shown a secular rise nationwide, such that EYSI has low likelihood of significant variance across states, even though the MYSI will penalize those with a past record of underperformance.

In short, the fastest way to ensure broad based visible HDI growth across Indian states is very straightforward - focus on income growth. This is also visible in India’s own performance on the Income Index - in the decade between 2015 and 2025 it has grown GDP PPP per capita from $5300, corresponding to the threshold between Low and Medium tiers, to $12000, well within the High tier.

An aggressive and unrelenting focus on income growth has the direct benefit of enabling and/or incentivizing improvement in the health and education related HDI parameters. Given the means, no one wants to be sick or uneducated - especially not if they can easily gain access to substantial incremental income by investing incrementally in better health and education.

Final Thoughts

The widely publicized 2023-24 UNDP HDI score for India is misleading. Its principal failure is unrepresentative data. It uses a questionable life expectancy metric three years below the officially reported trend line. Utilizing current data shows that India ranks in the Very High HDI tier for this metric.

On the other two metrics, it is yet to capture the growth in gross enrollment, which will take time to manifest itself in the expected years of schooling. The GNI per capita data is almost half the figure for 2025, due to the spectacular growth in purchasing power since 2021.

In effect, the UNDP HDI data can be charitably described as laggy and unrepresentative of the present day when it comes to data for India. While it might suit some political narratives, wholly unrepresentative metrics are also unusable for any real policy making effort - which depends on feedback about failures and successes.

The takeaway of this exercise is that India effectively has an HDI of approximately 0.715 - classifying as High tier -  today. Given their reporting lag, these figures will likely not be reported by the UNDP at least until 2026-27 or later. 

In effect, this shows that Indian HDI grew across more than a full tier worth (i.e. +0.100)  in 10 years, from 0.609 in 2015 to approximately 0.715 today - approximately the level of China in 2014. This improvement is on par with China improving its HDI from 0.588 to 0.699 between 2000 and 2010 - the most dramatic decadal growth they demonstrated.