MDRIA and a New Line of Defense for Sierra Leone's Extractives
  • Sierra Leone extractive revenue
  • low-capacity audit targeting
  • diamond rutile taxation
  • small economy revenue recovery
  • explainable audit selection
  • mining compliance West Africa
  • resource governance Sierra Leone

Sierra Leone's mineral wealth — diamonds, rutile, bauxite and iron ore — represents both its greatest economic hope and its most persistent revenue vulnerability. Under the Mines and Minerals Development Act 2022, the National Revenue Authority (NRA) faces the task of collecting royalties and taxes from complex operations where physical volumes, grades and transactions are easy to misreport and hard to verify.

01

Why every revenue dollar matters

For Sierra Leone, extractive revenue carries unusual weight. The mining economy is built around iron ore, diamonds, rutile and bauxite, alongside artisanal gold and diamond production. EITI's latest country data shows extractives contributing about 11% of government revenues and 61% of exports.

11%
of government revenues from extractives
61%
of exports from the extractive sector
$41M
extractive tax revenue in 2023

In 2023, extractive tax revenues reached about US$41 million, led by iron ore, diamonds and rutile. In a smaller economy with limited administrative capacity, Sierra Leone extractive revenue lost through under-reporting, mispricing or weak reconciliation can therefore have an outsized fiscal cost.

02

Designed for a different audit constraint

The Multi-Dimensional Revenue Intelligence Algorithm (MDRIA) is especially relevant to low-capacity audit targeting because it does not require a large archive of confirmed fraud cases. Instead, it asks whether a taxpayer's production, prices, costs, royalties, exports and physical activity can all be reconciled.

Auditors can see which constraint triggered concern and why a case moved up the risk ranking.

The framework compares company-controlled declarations with harder-to-manipulate references such as utility or fuel records, customs data, market prices and statutory obligations. The result is explainable audit selection.

03

From diamonds to rutile: following the physical story

That logic could be adapted across Sierra Leone's main minerals. For diamonds, reported production and exports could be reconciled with licensed activity, valuation records and Kimberley Process information. For rutile and iron ore, declared output could be tested against processing activity, fuel or electricity consumption, shipment volumes and export records.

Diamond, rutile & iron ore focus

High-value resources carry higher revenue risk.

Self-verifying tax reports

Reports that prove themselves against independent evidence.

Low-capacity audit targeting

Find the highest-risk cases first.

Partner-country import matching

Stops collusive or spread-out evasion.

Diamond rutile taxation then becomes part of a wider consistency test rather than a stand-alone calculation. A royalty may be arithmetically correct while still being calculated on an understated production or sales base.

04

Why multiple weak signals matter

The paper's central insight is that sophisticated evasion may be spread thin. A company could slightly understate output, reduce export price, inflate costs and adjust its royalty base without making any single number look extreme. MDRIA combines those weak inconsistencies instead of waiting for one large red flag.

A result from simulation, not Sierra Leonean data — in the paper's synthetic experiment, this multi-channel strategy was the easiest fraud type for the joint-consistency framework to detect. That result comes from simulation, so local validation would still be essential.
05

Doing more with limited audit capacity

This matters because small economy revenue recovery depends heavily on choosing the right cases. The framework ranks taxpayers under a fixed audit-capacity constraint, allowing scarce investigative resources to focus where several independent facts disagree.

It also avoids relying on a black-box score alone: each selection can be traced to production, pricing, cost, royalty, revenue-feasibility or mass-balance evidence. For mining compliance West Africa, that transparency matters because audit decisions must remain defensible to taxpayers and institutions.

06

Building on Sierra Leone’s existing data

Sierra Leone is not starting from zero. EITI reporting already reconciles company payments with government receipts, while the National Minerals Agency operates a digital mining cadaster and data repository. EITI has also highlighted gaps in the comprehensiveness of some production and export information, particularly around artisanal mining.

A stronger multi-source system could connect National Revenue Authority filings, National Minerals Agency production data, customs declarations, export valuations, royalty records and independent physical evidence.

07

A leaner line of defense

For resource governance in Sierra Leone, MDRIA would not replace auditors, investigations or due process. Its value would be narrower and practical: helping authorities decide where limited audit effort is most likely to matter.

By turning disconnected administrative records into explainable consistency checks, the framework could strengthen resource revenue protection without requiring a large, labelled fraud database or an opaque machine-learning model. For a smaller resource economy, doing more with less is part of protecting the public value of every ton, carat and dollar the extractive sector generates.

Better insights. Stronger compliance. More revenue for Sierra Leone.

Rank under a capacity constraint

Focus scarce investigative resources where independent facts disagree.

Trace every selection

Keep decisions defensible — each flag traces to specific evidence.

Combine weak signals

Catch spread-thin evasion that no single number would reveal.

Build on data already held

Connect NRA, NMA, customs and EITI records already inside government.

Collaboration

Collaborate with Maitras.ai

Revenue authorities, energy regulators and development institutions are seeking data-driven approaches to protect extractive revenue and strengthen fiscal transparency. Let's build physics-constrained audit intelligence together.

Start a conversation