Converging to Convergence

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/stata_convergence2/data/"
use "${BASE}main_data.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/stata_convergence2/data/"
df = pd.read_stata(BASE + "main_data.dta")

# load every dataset at once
files = ["main_data", "WDICountry"]
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 + "main_data.dta", "main_data.dta")
df, meta = pyreadstat.read_dta("main_data.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/stata_convergence2/data/"
df <- read_dta(paste0(BASE, "main_data.dta"))

Overview & sources

Companion data for a hands-on Stata tutorial that reproduces the main findings of Kremer, Willis & You (2021), “Converging to Convergence.” The analysis documents how the cross-country relationship between economic growth and initial income reversed — from divergence in the 1960s to unconditional convergence by the 2000s — and explains the shift through the convergence of growth correlates. It estimates beta-convergence (growth on initial log income) and sigma-convergence (the cross-country spread of log income), year-interacted rolling beta regressions, income-quartile and regional decompositions, and the omitted-variable-bias (OVB) identity that decomposes the gap between unconditional and conditional convergence as β − β* = δ × λ. The data are real: the authors’ replication panel combines Penn World Table 10.0 GDP with 50+ institutional, policy, geographic, and cultural correlates. The analysis is purely descriptive — it documents correlations and trends, not causal effects.

Data sources

Cite this data

Please cite this dataset as follows.

APA

Mendez, C. (2026). Converging to Convergence: Understanding the Main Ideas of the Convergence Literature [Data set]. https://carlos-mendez.org/post/stata_convergence2/

Kremer, M., Willis, J., & You, Y. (2021). Converging to Convergence. NBER Working Paper 29484. Barro, R., & Sala-i-Martin, X. (1992). Convergence. Journal of Political Economy, 100(2), 223-251. Feenstra, R. C., Inklaar, R., & Timmer, M. P. (2015). The Next Generation of the Penn World Table. American Economic Review, 105(10), 3150-3182.

BibTeX

@misc{mendez2026stataconvergence2,
  author       = {Mendez, Carlos},
  title        = {Converging to Convergence: Understanding the Main Ideas of the Convergence Literature},
  year         = {2026},
  howpublished = {\url{https://carlos-mendez.org/post/stata_convergence2/}},
  note         = {Data set}
}

@techreport{kremer2021converging,
  author      = {Kremer, Michael and Willis, Jack and You, Yang},
  title       = {Converging to Convergence},
  institution = {National Bureau of Economic Research},
  type        = {NBER Working Paper}, number = {29484}, year = {2021}
}
@article{barro1992convergence,
  author  = {Barro, Robert J. and Sala-i-Martin, Xavier},
  title   = {Convergence},
  journal = {Journal of Political Economy},
  volume  = {100}, number = {2}, pages = {223--251}, year = {1992}
}
@article{feenstra2015pwt,
  author  = {Feenstra, Robert C. and Inklaar, Robert and Timmer, Marcel P.},
  title   = {The Next Generation of the Penn World Table},
  journal = {American Economic Review},
  volume  = {105}, number = {10}, pages = {3150--3182}, year = {2015}
}

Variable explorer search & filter all 122 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

Convergence is measured on real GDP per capita. Let y_it be country i's GDP per capita in year t and g_{i,t→t+10} its 10-year forward annualized growth rate (the outcome loggdp_growth_10).

• Beta-convergence: g_{i,t→t+10} = α + β·log(y_it) + ε — a negative β means poorer countries grow faster (convergence).
• Trend in beta (year-interacted): g = β_t·log(y_it) + μ_t + ε, one β_t per year, with year fixed effects μ_t.
• Sigma-convergence: σ_t = SD_i[ log(y_it) ] — the cross-country standard deviation of log income each year; a narrowing series is sigma-convergence.
• Conditional convergence: g = α + β*·log(y_it) + λ·X_it + ε, where X is a growth correlate and β* is the convergence slope holding it fixed.
• Correlate-income slope: X_it = ν + δ·log(y_it) + u — how much richer countries have more of the correlate.
• OVB identity: β − β* = δ × λ — the gap between unconditional and conditional convergence equals the product of the correlate-income slope and the growth-regression slope (an exact algebraic identity).

Speed and half-life of convergence follow from β: the implied convergence speed is λ_conv = −ln(1 + β)/T and the half-life is ln(2)/λ_conv (here T = 10 years). Income variables are in natural logs (loggdp from PWT real GDP per capita); correlates are normalized by their 1985 cross-country standard deviation in the OVB exercises so coefficients are comparable across variables.

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