1. A Chiefly Banker moves in to corner GoldBod financing deal
The Chairman of LVSafrica is a traditional chief, Chair of state-owned enterprises, and the decades-long former CEO of a major multinational bank in Africa.
While many of us were shouting ourselves hoarse about GoldBod’s losses and whether GoldBod could live with those losses or not, he was busy in his exquisitely paneled huddle room with his sharpest analysts crafting a prospectus.
Some frantic TV studio debaters were trying very hard to sell the idea that GoldBod’s losses were inconsequential natural “costs of policy” and should be tolerated. But anyone who had seriously looked at the numbers and models knew that leaving things be was not a viable option. At some point, the Bank of Ghana would need to step aside, and not just because of the IMF. The Bank’s own risk analysts had concluded that it was overexposed.
The chiefly banker of course knew this. And he was prepared to offer a solution with a fat payoff for his firm.
His solution, ostensibly designed in partnership with the Ghana Association of Banks (GAB), is a plain-vanilla short-tenor commodity trade-finance deal: banks provide cash in cedis to aggregators who buy gold, GoldBod takes custody and exports, off-takers remit foreign exchange into an escrow account at the Bank of Ghana, and the pool revolves. Problem solved, no? No more noise about Bank of Ghana’s balance sheet being used for commodity speculation, or?
The problem is that the prospectus does not truly move GoldBod away from Bank of Ghana dependence. The Bank of Ghana is still placed at the centre of the structure as transaction obligor, escrow bank, foreign-exchange allocator, and source of prudential relief. The latter simply means that the Ghana Association of Banks and LVSafrica wants the Bank of Ghana to bend its rules for them so that they can chip in some cash for the gold trade and cream off the top. In substance, the proposal changes the form of central-bank support more than the fact of central-bank support.
My analysis of the offer concludes that it mainly solves a timing mismatch, but it does not address fundamental profitability or resilience problems. If GoldBod continues to buy in a politically charged market, continues to sell into concentrated external channels, insists on opaque discounts on the sell-side and premia on the buy-side, and continues to rely on policy privilege for economics, then private banks chipping in a cedi here and a cedi there for a juicy cut of the same margin merely spreads the pressure across the banking system instead of eliminating it.
The LVSafrica proposal is also pretty thin on detail. It is tenuous and hazy about facility size, legal basis for the Bank of Ghana obligor role, hard evidence of the rationality of the reserve-requirement and single-obligor waivers, collateral perfection, insurance, hedging, default waterfalls, buyer-concentration caps, sanctions controls, and enforceability of aggregator guarantees.
My base judgement is that the deal is very significant due to the eminence of the promoters and their excellent timing and astute political/political economy judgement. Thus, it deserves serious attention, but it is not ready for transactional primetime yet.
2. What the prospectus actually proposes
The prospectus in question is a concise non-binding proposal touting a “Revolving Structured Domestic Gold Trade Financing Syndication” in which domestic banks provide cedi liquidity for gold purchases, GoldBod acts as platform manager and operational gatekeeper, and the Bank of Ghana anchors settlement through an FX escrow structure. The core features are set out below.
| Feature | Prospectus Offering | Significance |
| Liquidity/cash source | Syndicate of Ghanaian banks | The facility seeks to move working-capital funding from the central bank to domestic financial institutions. |
| Beneficiaries | Licensed gold aggregators buying from small-scale miners | The actual cash recipient is the aggregator (just like the current model) rather than GoldBod |
| Gold flow | Aggregators buy gold, deliver to GoldBod for assay, custody, export prep, and onward export | GoldBod remains the operational choke-point of the trade. |
| Repayment source | Off-takers remit FX export proceeds into a Bank of Ghana escrow account | Repayment is tied to commodity sale proceeds rather than tax revenues or long-term amortisation. |
| Role of Bank of Ghana | Transaction obligor, escrow manager, and FX allocator | This is meant to be the real credit enhancer in the structure. |
| Requested waivers | Reserve requirement, capital allocation cost, and single-obligor constraints | A large part of the lender return is driven by regulatory privilege. The deal makers don’t trust their own margin play. |
| Tenor | Short-term cycle, typically 2 – 10 days; one-year revolving line | The self-liquidating nature is attractive, but actual realised cycle time is too rosy about risk. |
| Projected bank return | Illustrative annualised gain of 17.87% on a GH¢1 billion / 30-day example | The economics are sensitive to funding rates, reserve assumptions, and the FX spread. But the promoters offer no sensitivity analysis. |
3. Sexy moment; impeccable timing
GoldBod burnt through GH¢88.52 billion in cedi cash/liquidity in the first three quarters of 2025. But it delivered gold worth GH¢86.51 billion over that nine-month period. Annualised, that is roughly GH¢118.0 billion of flow used to buy ~110 tons of gold. In other words: this is not some small working-capital gig. We are talking about the biggest game in town.
At the same time, the IMF has been complaining about losses and the overbearing role of the central bank in foreign exchange (FX) markets. Over the three quarters under IMF review, GoldBod’s trading and related losses hit about $214 million. Whilst GoldBod declared operating profits, the Bank of Ghana as ultimate purse-holder had to swallow the losses. Obviously, the game couldn’t continue forever under the same rules.
Hence the LVSafrica offering.
4. LVSafrica deal logic
GoldBod’s working-capital need is fundamentally a cedi-timing problem: money must be paid out to aggregators and miners before foreign-exchange proceeds come back. A revolving, self-liquidating structure is therefore more suitable than a term loan or a budget transfer.
Second, the proposal situates repayment in the underlying asset-conversion cycle. That is exactly what good commodity trade finance tries to do. The trade pays down the trade.
Third, the use of an escrow account and delivery-linked drawdowns is on point. In commodity finance, transparency improves when money, goods, and receivables can be tracked in the same chain.
Fourth, the proposed short cycle of 2 – 10 day tenor is exactly the kind of story you want to tell bankers. The shorter the cycle, the lower the exposure to gold-price moves, buyer delays, and fraud drift. Bankers salivate at that kind of risk math.
Unfortunately, the proposal makes the Bank of Ghana the transaction obligor, the escrow bank, and the FX allocator. That means the central bank remains the key source of credit comfort. In plain language: commercial banks are being asked to provide liquidity because the central bank stands behind the exposure. But, remember, this is exactly the arrangement that has led to the controversy in the first place.
At first glance, there is a risk of papering over the main issue, wrapping nice cloaks around the problem, and proclaiming an innovative solution at hand.
On top of that, the deal requires the Bank of Ghana to grant participating banks special concessions in the form of reserve-requirement relief, zero-risk-type treatment, single-obligor relief, and preferential access to FX that participating banks can re-sell. These are all embedded options that come at a cost. Currently, the private banks and their deal broker want the central bank, i.e. the public, to pick up the tab.
There are other gaps and kinks.
The collateral story is too vague.
A bankable metals facility normally specifies title transfer, pledge perfection, custody controls, transit insurance, assignment of receivables, buyer eligibility criteria, events of default, and exception handling. The prospectus mentions generic controls, but says almost nothing about how those controls will convert to recoverable security.
The guarantee architecture is contradictory.
The prospectus claims the Bank of Ghana would be the obligor, but also says the Bank of Ghana would be “covered” by bank guarantees from aggregators. That is confusing. Ultimately, though, the question is whether one can trust the insurance sector in Ghana to bolster the kind of aggregators we have in the current GoldBod trade, who are all mostly unsophisticated players in the Ghanaian corporate terrain.
The proposal is missing a ring-fenced borrower or SPV.
The structure reads more like an operational flow diagram than a true financing perimeter. Without bankruptcy remoteness, contractual segregation, and a tightly drafted borrowing base, it is hard to see a ring-fenced asset pool.
5. Where does the value really sit?
Is privilege obscuring value creation?
The prospectus’ illustrative return does not come mainly from superior gold trading. It comes from a privileged FX spread plus prudential waivers. The proposal could thus feel attractive to banks even if the fundamental economics and public policy cushion don’t add up.
It naturally doesn’t tackle key sources of risk like overconcetration
As I have said in previous commentary, GoldBod’s trade is overly concentrated. Over the July to October 2025 period, 98.2% of GoldBod’s export value went to India and the UAE, with the top four buyers accounting for about 78.4% of value. This risk isn’t even mentioned in the prospectus even though it a major fat-tailed distribution.
This model is biased by the bull market.
World Gold Council data show 2025 was an extraordinary gold year, with 53 new all-time highs and an annual average LBMA PM price of US$3,431/oz. A structure developed under those conditions may look more resilient than it really is. A lower-price regime would pressure miner incentives, procurement volumes, working-capital turns, and the politics of paying premia. All at once. The prospectus provides little indication of how the structure would perform under stress.
The Prospectus could have borrowed sharper insights from elsewhere
The Bangko Sentral ng Pilipinas’ gold-buying framework operates through gold buying stations, tax incentives under Republic Act 11256, and a responsible-gold-sourcing regime aligned with LBMA expectations. This is a Sustainable formalization framework that banks can trust for the medium-term. Erecting a liquidity machine on top of the current GoldBod approach downplays the weaknesses in the value chain due to persistent informality.
6. Summarising key risks in the deal structure
As suggested above, the real risk picture is considerably wider than the prospectus allows, and some of the most important risks are latent rather than obvious.
| Risk | Type | Mechanism | Severity |
| Regulatory-waiver risk | Overt | The lender return model depends on reserve, capital, and single-obligor relief. If waivers are withheld, delayed, or narrowed, the economics weaken sharply. | High |
| Bank of Ghana obligor/legal-authority risk | Latent | The structure assumes the central bank can comfortably serve as transaction obligor without policy or program friction. That legal/policy basis is suspect. | High |
| Aggregator guarantee failure | Overt | The guarantee concept is mentioned, but the issuer, collateral, triggers, exclusions, and recovery process are unspecified. | High |
| Wrong-way banking-system risk | Latent | If the same banks are both liquidity providers and guarantee issuers (because the insurance sector lacks credibility for the exposure) or even if they are just systemically linked to both, the risk is being recycled inside the same financial system. | High |
| Buyer / off-taker concentration risk | Overt | A small number of India/UAE buyers dominate flows. One serious dispute, sanction event, banking cut-off, or settlement delay could choke the entire chain. | Very high |
| Logistics and geopolitical corridor risk | Latent | Current Gulf-area disruptions can lengthen air and sea transit times and complicate special-cargo movements, directly stretching the working-capital cycle. | Very high |
| Price/basis risk | Overt | Short tenor helps, but does not cancel the difference between procurement pricing and export realisation, especially if bonuses or discounts persist. | High |
| AML / ESG / responsible-sourcing risk | Overt | If provenance controls fail, buyer appetite, correspondent-bank tolerance, and reputational support can erode very quickly. | High |
| Macro / IMF-program tension | Latent | While the IMF program is ending, investors still continue to rely on the IMF for policy surveillance. If the proposal is judged to preserve quasi-fiscal central-bank exposure or deepen BoG’s market footprint, support from authorities and partners could weaken. | High |
7. The cycle rate deserves special attention
First, GoldBod’s working-capital need is extremely large relative to the illustrative model in the prospectus. At the annualised Q1 – Q3 2025 GoldBod / Bank of Ghana flow, a GH¢1 billion pool only does serious work if it turns very fast and very reliably.
Second, any extension in the cycle caused by logistics, assay disputes, export bottlenecks, or buyer delays causes the required pool size to jump sharply.
Third, even if the cedi bridge is solved, the structural leakages in the trade model discussed during the recent controversy (such as procurement incentives, discounts, freight, settlement basis, and policy interference) still remains. As should be evident from the table and chart below, realistic jolts to logistical turnaround have very severe effects on marginal profitability of the entire scheme.
| Average cycle (days) | Annual throughput from GH¢1bn pool (GH¢ bn) | Share of annualised 2025 Q1 – Q3 GoldBod/BoG flow | Implied pool size needed to match annualised flow (GH¢ bn) |
| 3 | 121.3 | 103% | 0.97 |
| 5 | 72.8 | 62% | 1.62 |
| 7 | 52.0 | 44% | 2.27 |
| 10 | 36.4 | 31% | 3.24 |
| 30 | 12.1 | 10% | 9.73 |
This is one of the proposal’s weakest points. The prospectus gives an illustrative GH¢1 billion return example, but it does not clearly state the target pool size required to sustain GoldBod at observed 2025 scale. Without that, banks are left to speculate whether they are evaluating a pilot, a bridging tranche, or a system-replacing facility. One presumes that clarifications are being left to the in-person meetings but our sources are unable to explain clearly how LVSafrica is addressing the issue.
8. Probing further into the returns story by digging into the cycle rate issue
The prospectus’ appendix provides a GH¢1 billion / 30-day illustrative bank-return model. Under that example, the marketed annualised return is 17.87%. But analytically that number rests on a small number of assumptions that may not hold steadily in practice.
The example assumes the Bank of Ghana sells FX to participating banks at 10.75 while the banks re-sell to clients at 11.00. That 25-pesewa spread is the primary earnings engine.
The example assumes an 11.5% funding / overnight opportunity cost and a 10% reserve-requirement saving. Yet the Bank of Ghana’s current monetary-policy stance has reserve requirements trending downwards past 8%.
The model also assumes the policy privileges are frictionless. In reality, insurers, lawyers, collateral managers, operations teams, and control functions all cost money. So does the pricing of any real guarantee. Those leakages are basically omitted.
| Scenario | Funding assumption | CRR | FX spread | Scenario-based annualised return |
| Prospectus assumptions | 11.5% | 10% | GHS 0.25/$ | 17.9% |
| CRR 8%, funding 15.9% | 15.9% | 8% | GHS 0.25/$ | 13.6% |
| Same, spread 15p | 15.9% | 8% | GHS 0.15/$ | 2.3% |
| Same, spread 10p | 15.9% | 8% | GHS 0.10/$ | -3.3% |
| Funding 18.0%, spread 25p | 18.0% | 8% | GHS 0.25/$ | 11.7% |
That sensitivity result is one of the most important metrics for analysts to track. With the prospectus’ own structure, the advertised return can collapse from the high teens to low single digits if the FX spread tightens. At a 15-pesewa spread, the illustrative annualised return falls to about 2.3%; at 10 pesewas, it turns negative under the same high-level funding assumptions.
9. Be sensitive to sensitivity
The prospectus could do with a lot more disclosure to assist in proper scenarios modelling. Below, I list some critical gaps.
• No stated target facility size or hard syndication amount.
• No disclosed legal opinion confirming the Bank of Ghana’s obligor role and the treatment of that exposure for prudential purposes.
• No signed or referenced waiver instruments for reserve requirement, capital allocation treatment, or single-obligor limits.
• No detailed security package: title chain, pledge perfection, receivables assignment, warehouse/custody protocol, or insolvency remoteness.
• No text around insurance, coverage schedule, loss payee language, or fidelity / transit / assay cover.
• No discussion of the costs and impacts of various hedging policies for gold-price or FX basis risk.
• No buyer-eligibility framework, concentration limits, sanctions screen, or payment-performance track record discussion.
• No waterfall for default, delayed settlement, assay dispute, or partial shortfall.
• No clear statement on who bears losses if GoldBod procurement pricing diverges from export realisation.
10. The Top Banks will force some serious reworking
LVSafrica has clearly positioned around a big hit. However, the sophisticated banks will likely stretch it hard to tighten the structure. A revamped model would preserve the valid intuition in the LVSafrica proposal while removing the ambiguities that make it weak and pretty much unbankable in its current form.
I suspect the following might come up as discussions move from exploratory to serious:
• Creating a ring-fenced borrowing-base SPV or trust structure instead of a loosely described “pool”.
• Acknowledging that the Bank of Ghana’s role might end up being limited to escrow, settlement, and policy oversight unless a fully explicit sovereign/central-bank guarantee is legally authorised and fiscally owned.
• Demands for some substantive reforms at the GoldBod may be made. For example, transparency around discounts, subsidies, and premia..
• Require perfected security over gold, receivables, and related documents, with independent collateral-management and reconciliation.
• Some bolder banks may even suggest concentration caps by off-taker, country, and aggregator. Though I doubt that the GoldBod would want to tie its hands.
• Using dynamic mark-to-market controls and margin calls for assay shortfalls or price movement.
• Requiring full insurance and named loss-payee arrangements.
• Publishing a pricing formula and standard fee stack so the system is not run on opaque discretion.
• Separate guarantee issuers from facility lenders where possible, or require hard collateralisation of guarantees.
• Provide an independent reporting pack: daily gold movement, outstanding cedi exposure, escrow balance, buyer ageing, exceptions, and realised trade economics.
• Stress-test the model under lower gold prices, longer cycle times, and Gulf/India corridor disruption.
Be all that is may, many of us will remain glued to our usual intel channels for more filla on how all this pans out eventually. Big money is about to shower into some chambers.
NOTE: Nothing in this essay is investment, legal, or regulatory advice.