Uganda's gold export levy: a lesson revisited

When Uganda imposed a substantial levy on gold exports in the early 2020s, declared volumes fell sharply and refiners suspended shipments. The measure was subsequently restructured. The metal did not stop moving; it stopped being declared in Uganda.

The speed of the response was itself instructive. Declared export volumes did not decline gradually as operators adjusted; they fell abruptly, consistent with a flow that had several readily available alternative routes and switched to them almost as soon as the new levy took effect, rather than a flow that was genuinely captive to Ugandan infrastructure.

Fiscal instruments applied to mobile, high-value, easily concealed commodities need to be set below the cost of evasion, including the risk premium. Above that threshold the base disappears and receipts fall despite a higher rate.

This is a specific instance of a much more general principle in tax design, sometimes summarised as the Laffer relationship, though the mechanism here is narrower and more mechanical than the broad macroeconomic version of that argument. It does not require any assumption about incentives to work or invest; it only requires that an alternative, undeclared export route exists and is cheaper than the new levy, which for gold moving through a porous regional border is almost always true above a fairly modest rate.

The same logic governs the current value-addition levies. Charges that make domestic refining marginally attractive will be paid. Charges that exceed the cost of routing through a neighbour will not, and will damage the traceability the policy was meant to build.

The clearest evidence that current policymakers have absorbed this lesson would be a stated methodology for how new levies are calibrated, ideally referencing an estimated cost of the smuggling alternative rather than simply a target revenue figure. Public commentary from the ministry so far has not disclosed such a methodology, which leaves open the possibility that future levy decisions repeat the earlier episode's mistake rather than learning from it.

A second useful marker would be a published, time-series comparison of declared export volumes against the point at which any levy changes were introduced. If Uganda genuinely internalised the earlier lesson, that data should show a levy structure that has been deliberately kept below the threshold at which declared volumes collapse, and the data to confirm or refute that exists inside customs and central bank records even if it is not yet public.