Four years ago Nigeria tried to wall crypto off from its banks. On 5 February 2021 the Central Bank of Nigeria ordered financial institutions to close the accounts of anyone dealing in virtual assets. That order is gone. What replaced it is harder and more interesting than a reversal, and on 12 August 2026 it got its first operational test.
The CBN opened applications for the second cohort of its Regulatory Sandbox Programme, and for the first time the intake carries a dedicated Virtual Asset Service Provider track. The VASP track covers stablecoins, payments and settlement, custody, and wallets, the plumbing of a digital-asset economy rather than the speculative surface of it. A second, non-VASP track admits firms using permissioned data sharing to improve credit, payments, and risk. Applications close on 31 August 2026.
Read as a headline, this is “Nigeria lets crypto back in.” Read structurally, it is something a lot of larger markets have not managed: a central bank and a securities regulator agreeing, in advance and in writing, on who owns which part of the problem.
The split is codified, not improvised
The division did not emerge from the sandbox. It was written into law a month earlier. On 17 July 2026 President Bola Tinubu signed the Presidential Executive Order on Virtual Assets Coordination, which draws a functional line through the asset class. The SEC regulates virtual assets that behave like securities, the investor-protection axis. The CBN regulates virtual assets used for payment, settlement, and custody, the systemic and monetary axis. Where an asset does not sit cleanly on either side, a new Virtual Asset Council, chaired by the CBN and seating the SEC, the Nigeria Revenue Service, the financial-intelligence unit, and the national security adviser, decides which regulator holds it.
That last clause matters more than the split itself. Plenty of jurisdictions have two agencies with overlapping claims on crypto. What they lack is a pre-agreed arbiter for the cases that fall between them. Nigeria built the referee before the match. The order also plants a Virtual Asset Office inside the CBN to run day-to-day coordination, licensing intake, and reporting between the agencies.
So the sandbox is not the moment the split was created. It is the moment the split has to survive contact with real firms.
A supervised pilot, not a press release
The CBN did not walk into this cold. Since 31 March 2026 it has run a live anti-money-laundering supervision pilot over six named operators: the naira stablecoin cNGN, the payment processors Flutterwave and Paystack, the remittance firm Juicyway, and the exchanges KoinKoin and KuCoin. Those firms file monthly compliance reports on a CBN template, engage directly with the financial-intelligence unit, and submit to reviews of onboarding, sanctions screening, transaction monitoring, and cross-border flow. Each has to show a credible plan for the Financial Action Task Force travel rule.
That pilot is the CBN learning to supervise entities it spent 2021 trying to expel. The sandbox widens the same muscle to any firm that applies before the end of August. The bank now has both a rulebook and a year of supervisory reps behind it.
Where the seam will strain
A functional split reads clean on a slide and frays at the edges, because real products do not respect the boundary. Stablecoins are the obvious pressure point. cNGN moves value like a payment instrument, which puts it under the CBN, but a yield-bearing or investment-wrapped stablecoin starts to look like a security, which pulls it toward the SEC. Custody is another seam: holding a payment token and holding a tokenized security are the same operation with different legal weight on top.
This is precisely why the Virtual Asset Council is the load-bearing part of the design. The test of Nigeria’s model is not whether the CBN can run a sandbox. It is how often the Council has to step in to assign a firm, and whether it does so fast enough that applicants are not left guessing which regulator’s capital and conduct rules apply to them. A division of labor that needs constant adjudication is not a division of labor. A council that never meets means the seam was drawn in the right place.
The application window compresses that question. Firms have until 31 August to self-select a track, which means each one is effectively making a first-pass jurisdictional call about its own product, with the Council as backstop.
The template question
Nigeria is not the largest crypto market to sort this out, but it is one of the most instructive, because it is doing the sorting deliberately rather than through enforcement. The United States spent years letting the boundary between its securities and commodities regulators get settled case by case in court. Nigeria’s bet is that an emerging market can skip the litigation phase by naming the referee up front.
If it holds, it becomes a portable design. A central bank takes the payments and stability axis, a markets regulator takes the securities axis, and a standing council resolves the overlap, all anchored to activity rather than to a firm’s label. That is a structure a Kenya or an Indonesia can copy without importing a decade of American case law. If it does not hold, the failure will be visible in the same place the design is strongest: the Council will either become a bottleneck or a rubber stamp, and the clean functional line will blur back into turf.
The reversal of the 2021 ban was completed quietly on 22 December 2023, when the CBN issued guidelines letting banks service licensed VASPs again. The interesting part was never whether Nigeria would let crypto back in. It was whether two regulators could carve the sector without fighting over it. The sandbox closes on 31 August. That is when the answer starts to arrive.
Discussion
Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.