Spatial Inequality and the Kuznets Curve

Downloads

Each dataset is available as a labeled Stata .dta and its source file.

⇩ Download all data (ZIP)stata_codebook.do

Run stata_codebook.do in Stata once to attach long-form per-variable notes to the .dta files.

Load directly in code

Every file loads straight from GitHub (raw URLs). Swap the file name to load any dataset.

Stata

* Stata 14+ : `use` reads an https URL directly
global BASE "https://raw.githubusercontent.com/cmg777/starter-academic-v501/master/content/post/r_kuznets/data/"
use "${BASE}sim_country_panel.dta", clear
describe
notes

Python

!pip install -q pyreadstat
import pandas as pd
BASE = "https://raw.githubusercontent.com/cmg777/starter-academic-v501/master/content/post/r_kuznets/data/"
df = pd.read_stata(BASE + "sim_country_panel.dta")

# load every dataset at once
files = ["sim_country_panel", "sim_regional_gdp"]
data = {f: pd.read_stata(BASE + f + ".dta") for f in files}

# pyreadstat (richest metadata) reads LOCAL files -> download first
import pyreadstat, urllib.request
urllib.request.urlretrieve(BASE + "sim_country_panel.dta", "sim_country_panel.dta")
df, meta = pyreadstat.read_dta("sim_country_panel.dta")

Copy and paste this snippet in Google Colab app. https://colab.research.google.com/notebooks/empty.ipynb

R

# R : haven::read_dta auto-downloads an https URL
library(haven)
BASE <- "https://raw.githubusercontent.com/cmg777/starter-academic-v501/master/content/post/r_kuznets/data/"
df <- read_dta(paste0(BASE, "sim_country_panel.dta"))

Overview & sources

Companion data for a hands-on R tutorial that replicates Lessmann (2014) on a fully synthetic dataset of regional GDP per capita for 56 countries over 1980–2009. The post measures spatial inequality with the population-weighted coefficient of variation (WCV) of regional income and estimates the spatial Kuznets curve — the hypothesis that regional inequality first rises then falls as development increases — using cross-section OLS, two-way fixed effects (fixest), and semiparametric methods (Robinson 1988, Baltagi–Li 2002). The inverted-U is robust across estimators; the high-income upturn (N-shape) appears only in cross-country comparisons, not within countries over time. The entire data-generating process is open and reproducible.

Data sources

Cite this data

Please cite this dataset as follows.

APA

Mendez, C. (2026). Spatial Inequality and the Kuznets Curve: Parametric and Semiparametric Estimates in R [Data set]. https://carlos-mendez.org/post/r_kuznets/

Lessmann, C. (2014). Spatial inequality and development — Is there an inverted-U relationship? Journal of Development Economics, 106, 35–51.

BibTeX

@misc{mendez2026rkuznets,
  author       = {Mendez, Carlos},
  title        = {Spatial Inequality and the Kuznets Curve: Parametric and Semiparametric Estimates in R},
  year         = {2026},
  howpublished = {\url{https://carlos-mendez.org/post/r_kuznets/}},
  note         = {Data set}
}

@article{lessmann2014spatial,
  author  = {Lessmann, Christian},
  title   = {Spatial inequality and development---Is there an inverted-U relationship?},
  journal = {Journal of Development Economics},
  volume  = {106}, pages = {35--51}, year = {2014}
}

Variable explorer search & filter all 22 variables

Type to filter by name or label, or use the chips to filter by type. Each row shows a mini distribution. Click a header to sort.

Cross-file variable index

Which file each variable appears in (● = present).

Construction & formulas

Inequality is measured within each country-year, across that country's regions, on regional GDP per capita y_j with population shares p_j and country mean ȳ.

• Weighted coefficient of variation (wcv): WCV = (1/ȳ) · [ Σ_j p_j (ȳ − y_j)² ]^(1/2) — the headline scale-free index.
• Unweighted CV (cv): standard deviation / mean of regional gdp_pc (robustness).
• Regional Gini (gini_reg): population-weighted Gini over sorted regional income (robustness).
• WCV excluding the capital (wcv_nocap): drop region 1, re-normalize shares, recompute WCV.

Synthetic data-generating process: for region j in country i, year t, y_ijt = country_mean · exp(δ_it · z_ij), where z_ij is a persistent regional position (a rich region stays rich) and δ_it (log-dispersion) follows a structural inverted-U in development. A time-invariant between-country cubic term yields the cross-section N-shape while the panel keeps a clean inverted-U. Polynomial terms lnGDP2/lnGDP3 are powers of lnGDP.

The datasets

Switch datasets with the tabs. Each shows the full variable dictionary plus a sortable statistics table with mini distributions and data coverage.

Expand all datasets expand to search (Ctrl/⌘+F) or print across all datasets

Known limitations & caveats

• Synthetic data. There is no real data behind this tutorial; results are internally consistent with the calibration but are not empirical evidence about real-world spatial inequality.
• Fragile N-shape. The high-income upturn (cubic term) appears with income in logs and weakens in levels; the robust finding is the inverted-U, not the N-shape.
• Within vs. between. The genuine N-shape is cross-sectional (between countries); the within-country cubic term is insignificant.
• Significance ≠ shape. All income terms can be significant yet the discriminant D = β₂² − 3·β₁·β₃ be negative or the turning points fall outside the observed range — always check D > 0 and that turning points lie in range.