sagegar/visa-approval-refusal-rates

Visa approval and refusal rates: Schengen consulates and US nationalities

Three government datasets, normalised across years and made usable. The numbers are not mine — they are the European Commission's and the US State Department's. What is mine is the reconciliation: the EU publishes one spreadsheet per year with country labels that drift between them, and the US publishes PDFs.

Maintained at visachances.com, which is built from these files.

What's here

file	rows	grain	period
schengen_visa_outcomes_by_consulate_2014_2025.csv	22,027	Schengen state × consulate × year	2014–2025
us_b_visa_refusal_rates_by_nationality_FY2006_FY2025.csv	3,577	nationality × fiscal year	FY2006–FY2025
us_niv_refusal_rates_by_visa_class_FY2013_FY2024.csv	926	visa class × fiscal year	FY2013–FY2024

The Schengen file is the interesting one. It covers 28 deciding states, 357 consulates in 192 countries, and it is as far as I know the only public dataset that crosses where an application was lodged with which state decided it. The US publishes nothing equivalent.

Read this before you use the Schengen file

Two things will lead you to publish something wrong.

1. consulate_country is not nationality

The source column is headed "Country where the consulate is located." It is where the application was lodged, not the applicant's citizenship.

For a country whose nationals need a visa the two are nearly the same thing — almost everyone applying at a Schengen consulate in India is an Indian national. For a visa-exempt country they are not remotely the same. The 2025 file records 1,988 decisions at Schengen consulates in Brazil and 2,160 in Denmark: absurd as national refusal rates for 200 million Brazilians, or for Danes who need no Schengen visa at all, and perfectly sensible as third-country nationals resident in those places.

This is why "Schengen refusal rates by nationality" tables built from this file tend to show Denmark, Austria, Germany and Sweden among the lowest — those rows are a handful of resident third-country nationals, not the country's citizens. If you aggregate this column, say "lodged in", not "from".

2. The refusal rate is not not_issued / applications

The European Commission's own definition, which not_issued_rate_pct reproduces:

not_issued_rate = not_issued / (uniform_visas_issued + ltv_visas_issued + not_issued)

Decisions in a given year include applications lodged late in the previous year and exclude withdrawn ones, so applications is the wrong denominator and understates the rate. The column here was checked against the Commission's own published rate to fifteen decimal places.

A worked example, India 2024: 165,266 / (936,748 + 2,949 + 165,266) = 14.9567%

And one caveat on interpretation

A consulate's refusal rate reflects who applies there as much as how it decides. Germany's Mumbai and Kolkata posts apply identical law to different applicant pools. Treat these as outcomes, not a leniency ranking.

Read this before you use the US files

The US publishes three mutually exclusive marginal tables, and the cross you probably want does not exist:

• refusal rates by nationality — B visitor visas only
• refusal rates by visa class — worldwide, all nationalities pooled
• refusal grounds (214(b), 221(g)) — neither by nationality nor by class

So there is no official US student or work visa refusal rate for any single country. Not for India, not for anyone. The statutory basis is 8 U.S.C. § 1202(f), a FOIA Exemption 3 withholding statute. If you find an "F-1 refusal rate for India" anywhere, it is derived, and whoever derived it should say so.

adjusted_refusal_rate_pct is the State Department's adjusted rate, which nets out refusals later overcome. Two values in the series are legitimately negative (Luxembourg FY2006, Kuwait FY2009) because overcomes exceeded refusals in the period. A printed 0.00 is genuinely ambiguous in the source — every application issued, or no applications at all — and should not be read as a 100% approval rate.

Provenance and verification

Sources, both primary:

• European Commission, Visa statistics for consulates — short-stay visas issued by Schengen countries
• US Department of State, Adjusted Refusal Rate – B-Visas by Nationality and Nonimmigrant Visa Workload by Visa Category — visa statistics

What was verified: 22 country-year cells checked against the source PDFs across six fiscal years, all matching including deliberately odd mixed-precision values; the FY2025 nationality table checked line-by-line against an independent published copy, matching on all 191 values; every class-level year reconciled against its source document's own printed grand total; Schengen totals reconciled against the Commission's published headline figures for 2019, 2023, 2024 and 2025 (2025: 12,201,026 applications, 14.65% not issued, against the Commission's published 14.6%).

What was corrected: country-label drift. Only 173 of 245 raw spellings appear in all years; casing flips from Title Case to UPPER at FY2015, and CHINA - MAINLAND/CHINA, MACAU/MACAU S. A. R. are the same places. Left untreated, any cross-year comparison silently drops rows. Aggregate rows (NON-NATIONALITY BASED ISSUANCES, UNKNOWN NATIONALITY OR STATELESS) are excluded — they are not countries and would corrupt any ranking.

Known limitation: the US tables were extracted from PDFs by a language model rather than a deterministic parser. The verification above is what stands behind them; if you find a discrepancy against the source, it is a bug and I want to hear about it.

Licence

CC-BY-4.0 for this compilation. The underlying sources are reusable in their own right — European Commission documents under Decision 2011/833/EU with attribution, US federal government works as public domain. Please cite the primary source alongside this dataset; my contribution is the normalisation, not the figures.

Citation

@misc{visachances_visa_outcomes,
  title  = {Visa approval and refusal rates: Schengen consulates and US nationalities},
  author = {VisaChances},
  year   = {2026},
  url    = {https://visachances.com/data.html},
  note   = {Compiled from European Commission consulate statistics and
            US Department of State visa statistics}
}