Does Education or Income Better Predict HDI? Where the Two HDI Components Diverge

What we examined

The Human Development Index (HDI) is a composite built from three components: a health index, an education index, and an income index. This analysis isolates the second and third of those — the Education Index and the Income Index — and asks how they relate to one another across countries in a single verified year.

The dataset is the same 2022 cross-section used across CivlensAI’s research: 18 countries matched on ISO codes, with the HDI, Education Index, and Income Index drawn from the UNDP Human Development Report data. All three readings for every country refer to 2022, so the components are comparable within each row and across the sample.

Because both components are part of the same composite, they are related to the HDI by construction, and to each other through it. The question here is narrower: within that shared structure, where do the two components sit furthest apart?

The components move together, but not identically

At the broadest level the two components rise together: in this sample the Pearson correlation between the Education Index and the Income Index is 0.96. High-education economies tend to be high-income economies, and vice versa.

But the two are not the same measure. The difference between them — the Education Index minus the Income Index — spans 0.145 points across the sample, from India at −0.080 to New Zealand at +0.065. That range is not trivial: it is roughly three times the width of a typical country-to-country gap within the balanced middle of the table.

Finding 1 — In four countries the income component reads above education

India, Mexico, Brazil, and Singapore are the only countries in the sample where the Income Index equals or exceeds the Education Index. India is the clear extreme: its income reading (0.640) sits 0.080 above its education reading (0.560) — the widest gap in either direction anywhere in the sample.

Notably, three of these four — India, Mexico, and Brazil — are also the three lowest-HDI economies in the sample. That makes the income-forward direction a feature of the sample’s lower-development economies, with Singapore the exception: a very high-HDI economy whose income component still edges out its education reading.

Finding 2 — The largest education-above-income gaps are high-HDI economies

At the other end, the four economies with the widest education-above-income gaps — New Zealand (+0.065), Australia (+0.055), the United Kingdom (+0.055), and South Korea (+0.055) — all have an HDI of at least 0.925. In each, the education reading sits more than 0.05 above the income reading.

This mirrors the first finding: the direction of the component gap is not random. The education-forward direction clusters among the highest-HDI economies in the sample, just as the income-forward direction clusters among the lowest.

Finding 3 — Nearly identical HDI, opposite component gaps

Singapore and New Zealand sit almost side by side on overall HDI — 0.939 and 0.937, a difference of 0.002 — yet they stand at the opposite ends of the component gap. In Singapore the income component (0.920) exceeds education (0.905); in New Zealand education (0.920) exceeds income (0.855) by the largest margin in the sample.

The example shows that two countries can be effectively indistinguishable on the composite while holding different positions on its parts. Reading only the overall HDI would hide that internal contrast entirely.

Why this does not establish causation

Everything here is a description of how two published indices sit relative to one another in one year for eighteen countries. The HDI contains both components by construction, so their gaps feed into the composite by design; nothing in this analysis identifies education, or income, as a driver of the other or of development.

A small cross-section can shift under different samples, and index values of this kind are produced with particular methodologies that can change between report editions. These observations are a reason to examine more data — not conclusions about what causes a country’s development profile.

Limitations

• Small sample: 18 economies, treated as illustrative rather than representative of all countries.
• Observational single-year cross-section (2022): no trends, no controls, and no basis for causal inference.
• Component dependence: both the Education and Income Indices are components of the HDI, so each is related to the composite and to the other by construction.
• UNDP index methodologies and reference data can change between Human Development Report editions; the values shown are those held in the CivlensAI dataset for 2022.
• Small component differences (0.01–0.02) may not be meaningful at this resolution; larger gaps are highlighted only where the data surface them.
• At this resolution, an education-above-income vs income-above-education reading describes the two published indices, not the relative importance of education or income for a country’s outcomes.

Data and Methodology

This analysis uses indicator data already presented on CivlensAI. How that data is collected, organized, and updated is described on the Methodology page.

Sources

• UNDP Human Development Report

Related Indicators