Financial inclusion and regional growth in Colombia

Abstract

Does financial inclusion predict the economic growth of Colombia's departments once you take out the national trends that move all of them at once? To answer it, the project downloads nineteen public sources with a verifiable manifest, resolves every series to the municipal code, models them as a star schema with vintages on dbt and DuckDB, and builds a financial-inclusion index by dimension with frozen, published weights. On top go two annual panels, department and municipality, an interactive atlas, and an econometric battery of two-way fixed effects, cross-sectional and spatial dependence diagnostics, four designs that do not rely on the index being exogenous, and inference built for 33 clusters. The answer is published with its specification, its N and its tests, whatever the sign.

Atlas

The index on the map, by department and by municipality

Three views of the same data. Flat, with the metrics around it; raised, to lift one department out and look at it; and municipalities, to go down to all 1,123 units. Pinning a department lets you descend to its municipalities from the map itself, and choosing a region closes the frame in on that selection while the neighbours bleed off the edge.

• A territory without colour is not a zero: it is a territory no supervised institution reported that year.
• Block height is constant and encodes nothing. A height that depended on the value would let the units in front hide the ones behind, and a unit's reading must not depend on where it falls on the map.
• The San Andrés archipelago is a chip drawn off scale: it lies 700 km offshore and covers 26 km², so inside the frame it shrank the map by a fifth to paint a dot.
• Two of the panel's 1,123 municipalities have no polygon in the 2024 national map: they appear in the ranking and the figures, but not on the map.

Data

Nineteen public sources, each download fingerprinted

The index

The assumption gets measured before the method is chosen

Results

With entity and time effects, the coefficient is zero

0.89

wild cluster bootstrap p

999 Rademacher draws with the null imposed over 33 departments

0.68

permutation placebo p

499 shuffles of the index within each year; the real coefficient sits in the middle of the cloud

2.46

Pesaran CD on the residuals

weak but present cross-sectional dependence (p = 0.014); hence Driscoll-Kraay alongside the cluster

• 1The three dimensions on their own, the specification in changes, CCE, SLX and the two alternative indices all give the same thing: zero.
• 2The only design with a signal is the shift-share. It survives initial income as a placebo, but initial urbanisation produces an equally significant slope (+0.051, p = 0.013): it captures that the more urban departments, which are also the more included ones, grew faster over the period. It is a differential slope, not an effect of the index, and it is published as such.
• 3The event study around 2020 cannot test pre-trends — the dependent variable starts in 2019 and the shock is 2020 — and it is published with that limitation written down.

Decisions

Every decision of substance is written before the code

The decisions that change a result — frequency, data model, index method, econometric design — go into a decision record with their alternatives, their cost and how to reverse them. Constants live in one place and no published figure exists without the test that holds it up.

Reproduce

One command brings the whole project up

1. uv sync

   Exact dependencies from the lock file. Nothing is installed outside it.

2. uv run iif acquire all

   Downloads the nineteen sources, verifies the count against each API and writes the manifest with sha256.

3. make check

   Ruff, 91 Python tests, 250 dbt models and tests on DuckDB, and the site render. It exits zero or there is no commit.

4. uv run iif econ

   Runs the full econometric battery in under half a minute and writes the file every figure on this page comes from.

5. uv run iif atlas

   Generates the geometry and series this map consumes, inside the 3 MB budget the contract sets.

The local and CI engine is DuckDB. BigQuery is kept as an alternative target with a per-query cost cap, and Snowflake as a later demonstration; nothing in the project depends on either staying alive.