Tax Lab · Macro-fiscal & revenue mobilisation
Domestic Revenue Mobilisation Decomposition
Uganda's net URA collections as a share of GDP, 2016/17 to 2024/25, decomposed by tax head. The gap-closure simulator further down is a clearly separate, illustrative layer.
The ratios and achievement rate are real, recomputed over one GDP series. The gap-closure simulator further down is illustrative.
What this shows
The Tax Lab catalogue describes this prototype as a Stochastic Frontier Analysis over a 40-country African panel (Battese & Coelli, 1995), estimating a revenue "frontier" and attributing Uganda's gap from it to policy, compliance and structural causes. That panel isn't shipped in this repo, so rather than fabricate one, this page leads with what is real: Uganda's net URA collections as a share of GDP, from the URA Data Book, decomposed by tax head, plus how often URA has met its own collection target. A second, clearly separate section lets you explore the prototype's own three-channel decomposition idea as an illustrative "what would it take" layer, not a finding.
The data
Real · Net URA / GDP by tax head
Hover a year. Stack of Net URA collections as a share of GDP, by tax head (URA Annual Data Book / UBOS, Tables 1.1.10, 1.1.11, 1.1.13, FY2016/17-FY2024/25). "Other domestic" is the residual between total domestic taxes and the four named heads; the Data Book doesn't break out every minor head in this table. Every ratio is recomputed over one GDP series. GDP: Uganda Bureau of Statistics (UBOS), Annual GDP, June 2026 release (FY2025/26 publication tables), current prices, base year 2016/17.
Illustrative, not a forecast
Choose how many percentage points of GDP you'd want to add to net URA collections, over how many years, and how you'd split credit for it across three channels. The split below starts at the prototype's own stated default (45% policy, 35% compliance, 20% structural); move the sliders and it stays normalised to the gap you chose.
Where the real data comes from
URA Annual Data Book (Tables 1.1.10, 1.1.11, 1.1.13), via the compiled workbook; ratios recomputed by the Fiscal Intelligence and Reform Lab over one GDP series. GDP: Uganda Bureau of Statistics (UBOS), Annual GDP, June 2026 release (FY2025/26 publication tables), current prices, base year 2016/17.
Limitations
- The headline ratio is net URA collections over GDP, not "tax revenue over GDP": net collections may include non-tax revenue that URA collects.
- Every ratio uses one GDP series, UBOS's June 2026 release. The Data Book printed some ratios on older GDP figures (for FY2024/25, 13.98% against 13.88% here), and its FY2023/24 achievement rate divides gross collections by the net target (95.62% against 93.44% here). Both are documented in the Lab's evidence base.
- "Other domestic" in the stacked chart is a residual, not a named tax head the Data Book itself reports.
- The gap-closure simulator is a straight-line illustration, not a projection: it has no model of how policy, enforcement or growth actually interact.
- The policy/compliance/structural split is the prototype's own stated default weighting, not estimated from Uganda data; the sliders let you see how sensitive the split is to that assumption, not a better estimate of it.
- The real Stochastic Frontier model (African panel, Battese & Coelli specification) is not implemented here and would need the ICTD/WEO/WDI panel the prototype was designed to run on.
Sources: URA Annual Data Book, Tables 1.1.10, 1.1.11, 1.1.13; URA Annual Revenue Reports FY2019/20 and FY2020/21 (targets); Uganda Bureau of Statistics (UBOS), Annual GDP, June 2026 release. Catalogue method references: Battese & Coelli (1995), Empirical Economics 20(2); Fenochietto & Pessino (2013), IMF WP/13/244; Medina & Schneider (2018), IMF WP/18/17; ICTD Government Revenue Dataset.
The story built on this model, with the plans' own targets, the record, the documented measures and oil: The Road to a 20% Tax-to-GDP Ratio.