Open data: global wealth distribution by band from the Credit Suisse and UBS Global Wealth Reports, 2010 to 2023 (Country)

Rows: 20
Columns: 8

Wealth shares across four wealth bands for the global adult population, plus mean, median, and Gini figures for 20 countries. Free to download at datahub.io. Widely used in inequality research and public communication.

AI analysis

Dataset Analysis: Global Wealth Distribution by Country (2023)

1. Dataset Overview and Research Value

This dataset draws from the Credit Suisse and UBS Global Wealth Reports, one of the most authoritative and widely cited sources in inequality research. It captures a cross-sectional snapshot of wealth distribution across 20 countries for the single year 2023, combining aggregate metrics (mean wealth, total wealth) with distributional measures (median wealth, Gini coefficient). No source URL was provided in the metadata, but the Credit Suisse/UBS lineage gives this data strong credibility for academic and policy applications. Its primary value lies in enabling between-country inequality comparisons — particularly the contrast between mean and median wealth, which reveals the degree to which wealth concentration distorts average figures. The Gini column makes this dataset immediately actionable for correlation studies without requiring additional preprocessing, which is relatively rare for wealth data at this level of aggregation.

2. Data Quality Observations

Data quality here is notably clean for a 20-country cross-sectional file. Total null cells across all 8 columns is zero, which eliminates a common pain point in cross-dataset joins. No type mismatches are flagged, and all numeric columns resolve to appropriate types (Decimal or Integer as expected). The one area deserving scrutiny is the time coverage: the join-key range spans only a single date point (2023-01-01 to 2023-01-01), meaning the year column carries zero analytical variance — it is uniform across all 20 rows (distinct=1). This makes year effectively a static label rather than a usable dimension for temporal trend analysis within this file alone. Duplicate row counts are pending Phase D Parquet recomputation, so that metric should be verified before any merge operation. Given only 20 rows, even a single duplicate would represent a 5% data integrity issue worth investigating.

3. Key Column Distributions

The most analytically interesting columns are adultsmillions, medianwealthusd, and gini, each revealing different structural features of the data:

- adultsmillions is highly right-skewed (skew = 2.75), with a mean of 156M but a median of only 47.25M, and 3 flagged outliers — almost certainly reflecting populous nations like China, India, and the United States dwarfing the rest. The interquartile range (18.5M–111.3M) confirms most countries in the sample are mid-sized economies. - medianwealthusd shows a mean of ~$98,575 versus a median of ~$99,626 — unusually symmetric for wealth data (skew = 0.467) — but the 1 flagged outlier likely anchors one extreme (possibly a very low-income country with median wealth near the $2,840 minimum). The $56,868–$134,618 IQR suggests a meaningful spread across the middle 50% of countries. - gini ranges from 64.4 to 89.0 with a mean of 75.9 and relatively low spread (σ ≈ 7.7), indicating that all countries in this sample exhibit high wealth inequality by global standards — this is consistent with wealth Ginis being structurally higher than income Ginis. The moderate skew (0.422) suggests a slight pull toward higher-inequality countries. - totalwealthusdbillions is the most extreme in shape (skew = 3.14, 2 outliers), with a max of $145,800B versus a median of only $9,050B — a nearly 16× difference — reinforcing that total national wealth is dominated by a small number of very large economies.

4. Recommended Join Key Columns

For cross-dataset correlation, the natural join strategy depends on the target dataset:

| Join Key | Best Used When | |---|---| | countrycode (ISO standard) | Joining to macroeconomic, HDI, or governance datasets — most reliable for avoiding name-matching errors | | country (string) | Useful as a human-readable fallback, but normalize casing/spelling before fuzzy joins | | year + countrycode (composite) | Essential if joining to multi-year panel datasets to avoid row inflation |

Given that year is constant (2023 only), any join to a time-series dataset should filter the partner dataset to 2023 observations first. countrycode is the strongly preferred primary key due to its distinct=20 cardinality matching the row count exactly — a clean 1:1 mapping with no ambiguity.

5. Recommended Pairing Datasets for Correlation Discovery

Several dataset categories would combine powerfully with this wealth distribution snapshot:

- World Bank Income & GDP Data (2023) — Correlating meanwealthusd and gini against GDP per capita or GNI would test whether national income levels predict wealth concentration patterns differently than income inequality - UNDP Human Development Index (HDI) — Pairing medianwealthusd (a better welfare proxy than mean) with HDI scores, education indices, or life expectancy could reveal whether wealth distribution aligns with broader human development outcomes - Tax Revenue & Policy Datasets (OECD) — Testing whether countries with higher wealth taxes or inheritance taxes show lower Gini scores would be a high-value policy correlation - Financial Inclusion Data (Global Findex, World Bank) — medianwealthusd against bank account ownership rates or credit access could illuminate how financial system depth relates to wealth accumulation at the median - Earlier Credit Suisse/UBS Reports (2010–2022) — The dataset description notes coverage from 2010–2023; joining this 2023 file to prior years using countrycode + year would unlock trend analysis of Gini trajectories and wealth growth rates, which is arguably the highest-value use of this source series

⚠️ Analytical caution: With only 20 country observations, any correlation coefficients computed directly on this file will have very limited statistical power. Treat findings as directional and hypothesis-generating rather than confirmatory without broader country coverage.

Columns

year (date)
country (string)
country_code (string)
adults_millions (decimal)
mean_wealth_usd (integer)
median_wealth_usd (integer)
total_wealth_usd_billions (decimal)
gini (decimal)