- within Technology and Cannabis & Hemp topic(s)
A Gauteng High Court judgment handed down on 1 June 2026 concludes that Bitcoin is both “capital” and “money” under South Africa’s Exchange Control Regulations. The capital finding is defensible. The money finding is not. And the court’s treatment of what “export” actually means on a distributed ledger leaves the most difficult question unanswered.
Case No. 2022-029979·Wilson J, Gauteng Division, Johannesburg,·Decided 1 June 2026
I. Background
Between January 2018 and March 2020, the first applicant used two Luno accounts — one his own, one belonging to his co-applicant — to purchase just under 1,680 Bitcoin in South Africa, worth approximately R182 million. The Bitcoin was then transferred to wallets held on cryptocurrency exchanges registered outside South Africa. The Reserve Bank treated this as the unlawful export of capital without Treasury permission, and the Deputy Governor ordered forfeiture of approximately R6 million in Bitcoin and fiat assets standing to the applicants’ credit.
The applicants sought to review and set aside the forfeiture orders, principally on the basis that the Exchange Control Regulations, 1961 do not apply to cryptocurrency at all. Wilson J dismissed the application, finding that Bitcoin is “capital” — and indeed “money” — for the purposes of the Regulations, and that transferring it to foreign exchange wallets constituted an “export” in breach of Regulation 10(1)(c).
The judgment is also notable for expressly departing from an earlier decision of the same division — Standard Bank of South Africa v South African Reserve Bank 2025 (5) SA 289 (GP) — in which Motha J had reached the opposite conclusion.
II. The Capital Finding: Sound
The court’s conclusion that Bitcoin is “capital” under Regulation 10(1)(c) is the strongest part of the judgment, and it is largely unassailable. Following Oilwell (Pty) Ltd v Protec International Ltd 2011 (4) SA 394 (SCA), the court defines capital in its financial sense: a financial asset capable of holding value or being used as a medium of exchange. Bitcoin plainly satisfies both limbs. It can be bought and sold for rand. It is held in the expectation of appreciation. It is, in the language of the regulations, a financial capital asset.
The purposive argument is equally compelling. If Bitcoin falls outside the Regulations, exchange controls become trivially circumventable. Any person wishing to move value offshore need only convert rand to Bitcoin on a local exchange and transfer it to a foreign wallet. The entire architecture of capital controls would be defeated by the availability of a domestic smartphone app. On any principled reading, that cannot have been the intended effect of the Currency and Exchanges Act.
“Courts should be careful not to ascribe unusual or irreducibly exotic properties to phenomena which, though novel and perhaps unique in some respects, exhibit precisely the attributes an enactment is intended to regulate.”
The court’s rebuke of what it calls “magical thinking” — the tendency to treat Bitcoin’s technological novelty as a legal shield — is pointed and, on the capital question, well-aimed.
III. The Money Finding: Questionable
The court’s analysis becomes considerably less persuasive when it turns to the question of whether Bitcoin is “money.” The Regulations define money to include “any bill of exchange or other negotiable instrument,” and the court concludes that Bitcoin qualifies as a negotiable instrument because it is a right to a specified sum exchangeable for fiat currency and other things of value.
This characterisation is strained. A negotiable instrument is a written document that unconditionally promises or orders payment of a fixed sum, by or to a specifically identified or identifiable party. Bitcoin is none of those things. It is a cryptographic record of entitlement to a quantity of a scarce digital asset. There is no issuer, no obligor, no fixed sum, and no written promise. To describe a UTXO — an unspent transaction output on the Bitcoin blockchain — as a negotiable instrument requires the concept to do more work than it was designed to do.
More fundamentally, the court conflates money-like utility with money as a legal and economic category. The classic tripartite definition of money requires three attributes: medium of exchange, unit of account, and store of value. Bitcoin functions as a medium of exchange in limited contexts, and as a store of value in a manner analogous to gold. But it fails the unit of account test almost entirely. Its price volatility is so pronounced that virtually no merchant prices goods or services in Bitcoin natively. Those who nominally “accept” Bitcoin typically do so through payment processors that convert it to fiat at the point of sale. Bitcoin is tolerated as a settlement mechanism, not adopted as a monetary denominator.
The court’s observation that Bitcoin can be used to pay for goods and services therefore proves less than it claims. Atticus Finch, in Harper Lee’s To Kill a Mockingbird, accepted payment in vegetables from clients who could not afford his fees. That did not make turnips legal tender, or money, or a negotiable instrument. The fact that something is accepted in exchange tells you only that the parties agreed to treat it as having value in a particular transaction. It tells you nothing about whether the thing accepted meets the legal or monetary definition of money.
IV. The Export Analysis: Technically Undercooked
The most significant analytical gap in the judgment lies in its treatment of “export.” The court holds that transferring Bitcoin to a wallet on a foreign-registered exchange is self-evidently an export: “once the Bitcoin was placed beyond the Reserve Bank’s jurisdiction, it was exported.” The court supports this with an ATM analogy — a person drawing on a UK bank account at a Johannesburg ATM imports money; a person drawing on a South African account at a London ATM exports it.
The analogy is intuitively appealing but technically inapt, and it conceals a distinction the judgment never squarely addresses.
A bank account is the liability of a legal entity incorporated and regulated in a specific jurisdiction. Its location, for regulatory purposes, is determinate. A Bitcoin UTXO is not the liability of anyone. It is an entry on a distributed ledger — the blockchain — which is simultaneously replicated across thousands of independent nodes worldwide, many of them physically located within South Africa. Bitcoin does not exist in any single place. It exists as a consensus state of a global, jurisdictionally agnostic network.
This matters because the court’s export analysis elides two factually distinct scenarios. In the first, a South African resident converts rand to Bitcoin on a local exchange, then transfers the Bitcoin to a foreign wallet. In the second, a resident transfers rand directly to a foreign exchange, purchases Bitcoin offshore, then moves the Bitcoin between wallets or to cold storage. In the second scenario, the export event — the moment at which value leaves South Africa — is the rand transfer, not the subsequent movement of Bitcoin. When Bitcoin “moves” between wallets, what actually occurs is that the blockchain’s distributed ledger is updated: a cryptographic key changes hands, and a new address becomes associated with a particular quantity of Bitcoin. No thing crosses a border, because the ledger has no geographic location.
The judgment does not distinguish between these scenarios. It treats the transfer of Bitcoin to a foreign-registered exchange wallet as sufficient, in itself, to constitute export. But the registration jurisdiction of an exchange tells you where the exchange’s corporate entity is incorporated — not where the Bitcoin “is,” because Bitcoin is not “anywhere” in the physical sense the Regulations assume.
When Bitcoin “moves” between wallets, nothing crosses a border. The ledger has no geographic location. The court’s export analysis assumes a physicality that the underlying technology does not possess.
A more rigorous judgment might have avoided these difficulties by reasoning purposively rather than literally. The practical effect of the transactions was to remove R182 million in value from the Reserve Bank’s supervisory reach. Whatever the mechanics, the economic substance was a capital outflow. That argument — grounded in effect rather than in a geographic conception of where Bitcoin “is” — would have been more technically defensible and would not have required the court to map 1960s regulatory language onto distributed ledger architecture it was never designed to describe.
V. Bitcoin as Commodity: The Direction of Travel
Viewed against international developments, the judgment’s money finding is swimming against the tide.
The price dynamics of Bitcoin are instructive. Empirical analysis consistently shows that Bitcoin’s price is driven primarily by two factors: liquidity conditions and network effects. When central banks expand their balance sheets — when they create more money — the price of scarce assets rises. Bitcoin, with its mathematically fixed supply of 21 million coins, is among the most inelastic assets in existence. Its appreciation during periods of monetary expansion is not coincidental. It is a predictable consequence of scarcity in the face of currency debasement. This is not the behaviour of money. It is the behaviour of a commodity — more specifically, of a monetary commodity akin to gold.
Money, by design, is elastic. Central banks exist precisely to expand and contract its supply in response to economic conditions. Bitcoin’s supply cannot be altered. It is constitutionally fixed in the protocol. In this narrow but important respect, Bitcoin is the antithesis of fiat money: it is used, as gold is used, to preserve value against the erosion of purchasing power that monetary expansion produces.
United States legislators have recognised this. The Commodity Futures Trading Commission has treated Bitcoin as a commodity since at least 2015. The GENIUS Act (Guiding and Establishing National Innovation for US Stablecoins), passed in 2025, draws a sharp distinction between dollar-pegged payment stablecoins — regulated as money-adjacent instruments — and Bitcoin, which is explicitly excluded from the stablecoin framework. The legislative logic is telling: Congress was acknowledging that Bitcoin is not designed to function as a payment medium in the way a stablecoin is.
The CLARITY Act (presently being debated at Congress), representing the most comprehensive attempt at digital asset classification in the US, goes further. Its central innovation is a decentralisation test: an asset issued or operated without meaningful central control falls under CFTC commodity jurisdiction, not SEC securities jurisdiction. Bitcoin passes this test unambiguously. No person or entity controls its issuance, its protocol, or its monetary policy. Under the CLARITY framework, Bitcoin is the paradigm case of a digital commodity — fungible, scarce, issuerless, and priced by global supply and demand.
The combined direction of US legislation is therefore: stablecoins are regulated as money; Bitcoin is regulated as a commodity; other tokens are assessed on a spectrum. This taxonomy is both more technically accurate and more jurisprudentially coherent than the South African court’s approach, which forces Bitcoin into a 1961 definition of “money” that was drafted during an era of Bretton Woods fixed exchange rates.
VI. Conclusion
The Wilson J judgment reaches a defensible outcome. The transactions described — the funnelling of R182 million in Bitcoin through two accounts to foreign exchange wallets, without Treasury permission — are precisely the kind of capital flows that exchange controls exist to regulate. The capital finding, and the dismissal of the procedural fairness complaint, are both well-reasoned.
But the reasoning that supports the money finding, and the analysis of what constitutes an “export” on a distributed ledger, deserve scrutiny. The court dismisses the technological architecture of Bitcoin as irrelevant novelty. In doing so, it avoids the harder questions: Where, precisely, does Bitcoin exist? At what moment, exactly, does value leave South Africa? What is the export event when the asset being “moved” is an entry on a jurisdiction-agnostic global ledger?
The corrective to magical thinking about technology is not to ignore the technology. It is to understand it well enough to reason about it accurately. As digital assets become an increasingly significant feature of cross-border capital flows, South African courts will need to develop a more granular analytical framework — one that can distinguish between the export of fiat, the conversion of fiat to a digital commodity, and the subsequent movement of that commodity between cryptographic addresses on a global ledger. The present judgment conflates all three. Future cases will need to disentangle them.
Mangundhla and Another v South African Reserve Bank and Others, Case No. 2022-029979, Gauteng Division, Johannesburg. Wilson J. Heard 11 May 2026, decided 1 June 2026. Reportable. See also Standard Bank of South Africa v South African Reserve Bank 2025 (5) SA 289 (GP) (expressly departed from). Exchange Control Regulations, 1961; Currency and Exchanges Act 9 of 1933.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
[View Source]