PROPOSITION 1: A Ponzi scheme may exist if an economy has a large public sector (” is bounded below), and the assets of the state could be used for a bailout ($ is bounded below), and the probability of early termination of the Ponzi scheme by a regulator is low (2 is bounded above), and there is inexpensive access to citizens through the mass-media (c is bounded above), and there are no severe penalties on promoters of Ponzi schemes (d is bounded above).
PROPOSITION 2: A Ponzi scheme will exist even under partial bailout, if the condition in Proposition 1 holds and the probability of bailout, $, is higher than (1-n*), where n* is the critical fraction of citizens that are required to be involved for there to exist the possibility of a bailout.
—- Utpal Bhattacharya, “The Optimal Design of Ponzi Schemes in Finite Economies”, 2003.

The smart money is on the prediction that the more sophisticated regulatory frameworks around the world shall tend to balance technology growth and privacy protection if they are to retain their political legitimacy in an environment where both consumer rights and economic competitiveness have attained nearly equal status in policy debates around the word.

Skilled regulators have already begun to justify new reforms on the basis of privacy measures stimulating considerable technology progress.

Consequently, the growing concerns of consumers about the abuse of their personal data and the misuse of targeting algorithms to interfere with their decision-making autonomy are spurring some of the most fascinating work in the platform architecture design space today. A broad range of blockchain applications, for instance, is now anchored to the premise of facilitating greater user control over their own data, and provisioning of this data to service providers based solely on the wishes of the data owners.

Savvy governments have recognised this development and have begun developing regulatory frameworks that focus more on rewarding creative privacy management rather than stymieing novel business models and technologies based on some misguided, precautionary, principles. Others are just starting to align with the times.

Consider, for instance, Costa Rica’s Executive Decree # JP-40008. Enacted in December 2016 to amend extant provisions on privacy, the subsidiary legislation considerably transformed what, in the beginning, had been a wholesale, “precautionary approach”, regime into an innovation-compatible system of rules specifically designed to facilitate investments, business, and technology development in the data-rich arts and sciences. How does decree JP-40008 achieve these goals precisely?

Firstly, it retires the provision in Executive Decree # 37544-JP, an annex to primary Law # 8968, which had introduced a highly restrictive requirement for the “registration” of databases, registers, and other data repositories with the main data ombudsman in the country, the PRODHAB. Instead, it calls for the vetting of the security protocols employed to safeguard such data repositories against malicious breaches or inadvertent disclosures of personal data.

Furthermore, regulated financial institutions in the Central American country are exempt from the requirement of database registration with PRODHAB. The dynamics of inter-party certification in the financial industry, whereby such security and privacy certification is very often a prerequisite for interoperability (PCI-DSS plus being an obvious example), already delivers a higher standard for personal data protection in a more efficient and decentralised manner than can be achieved by most purely government-managed regimes.

The amendments also take into account the reality of cross-border data movements within federated entities by focusing on systematic compliance and downplaying overplayed concerns around jurisdictional fragmentation. The mere act of data crossing a border does not necessarily invoke jurisdictional issues if the technology platform observes uniform standards that may be higher than domestic requirements. The ability to investigate claims of abuse in electronic systems is rarely hindered, in practice, by such jurisdictional fragmentation, yet policymaking on “data sovereignty” and “data domiciliary” considerations frequently operates on unscientific notions that suggest that physical borders are determinate.

Costa Rica’s focus on ensuring that the country’s Data Protection Law evolves to reflect the growing appreciation of its technocrats for “embedded regulation” vindicates the hope that fast-paced technology progress can be aligned with pro-privacy regulatory regimes.

Embedded regulations seek to strengthen industry standards and promote cross-network accountability among industry actors in a relatively more decentralised fashion. Thus, whereas in the previous regime, “written individual consent” was required, the amendments now enable the use of digital assent, bringing the process more closely in line with the fast-growing trend of “e-signature management as a service”. The pace of innovation in the e-signature management space is such that the cost of complying with “individual consent” shall continue to drop dramatically without sacrificing the quality of compliance.

The experience of Singapore is also instructive in clarifying this “embedment” notion of weaving of regulations into the fabric of a country’s technology enterprise culture.

In Singapore, the Personal Data Protection Commission (PDPC) sees itself as a “capacity building” institution mandated to bear a significant portion of the costs and capacity burden of transitioning business, particularly small and upstart businesses, from complacency and ineptitude to readiness and vigilance. PDPC strives to transform enterprises of varying levels of sophistication into data-savvy operators equipped with the latest tools for complying with the law, whilst contributing at the same time to the tiny entrepot’s declared vision of becoming the world’s “data hub”.

Singapore’s government has invested in a significant range of compliance tools for seamless compliance tracking and reporting so that small businesses seeking to create disruptive technologies would not be distracted from that state-sanctioned mission.

This does not mean however, as it might seem at first, that consumer needs and rights have been deprioritised. On the contrary, the country is convinced that improved privacy protection is a technical-investment public good that must be addressed as a baseline for its technology industry to leverage for leadership.

The government of Singapore, in the context of dynamic privacy protection, refers to the “embedded regulations” notion used above as, “data protection by design”. This language has become popular in recent years within stringent regimes, but the assumption in such regimes has usually been that businesses are responsible for rebuilding critical infrastructure in order to comply. The Singaporean government, on the other hand, takes the view that this is best achieved through the cultivation of an “ecosystem of trust”, and that the key role of the public sector is not primarily that of a police service, enforcing aloof laws on a suspicious crowd of businesses, but that of an investor safeguarding a key resource: trust.

The Singaporean government’s posture on this matter is summed up in this quote from the country’s data ombudsman:

“The key challenge lies in enabling the use and disclosure of data to support the progress of technology and innovation, whilst protecting personally identifiable information, to allay privacy concerns.”[1]

By highlighting the fact of businesses confronting considerable reputational and business-disruption challenges when data is mishandled, Singapore’s privacy regulators have succeeded in driving consensus on a baseline of “data hygiene and ethics” that fosters collective action. Such action when backed by public investments contributes to advancing the state’s preferred motif of an “ecosystem of trust” beyond rhetoric into substantive interventions in critical data governance areas such as disputes resolution, advanced notifications of disclosure, profile reviews, and aggregation.

Whilst many countries focus on writing laws that merely heighten the risk barriers for legitimate enterprises but do nothing in facilitating the identification and penalisation of rogue operators, Singapore prefers a broad principles-based regime coupled with an active, co-investing, regulator that is respected by consumers for operating a transparent and highly communicative process, and trusted by businesses for a pro-innovation mindset that welcomes joint exploration of how to advance risk-fraught emerging disciplines such as big data, supervised machine learning on live diagnostic data, and behaviour profiling.

It is too early to conclusively judge whether or not role modelling in the international community will be sophisticated enough for the experiences of the likes of Singapore and, even, Costa Rica to become yardsticks of emulation. But with the heating up of competition in the machine learning space, it is very likely that international data treaties among like-minded countries shall in due course begin to drive the formation of “smart country leagues” akin to “free trade areas”. Data treaties should, in places like East Africa, prevent unnecessary replication of infrastructure whilst at the same time addressing concerns about “sovereign” data control. Should this happen, the world is likely to witness some short-term schism in the trajectory of data innovation, a veritable new digital divide between countries in pro-innovation “data leagues”, and those locked out due to incompatible privacy and data protection regimes.

In the long-term, however, the sense that only the rapid advancement of above-board, and done in the open, technology can safeguard consumers and citizens from powerful, malicious, actors, who do not give a toss about privacy, is likely to prompt an overall race to the top.

[1] Leong Keng Thai quoted on the website of the Info-communications Media Development Authority in an article titled: “Balancing Innovation & Personal Data Protection”, posted on 3rd November, 2017.l

When the Menzgold issues started to garner public traction, not many people thought it was interesting enough to spend any time analysing them. A few of us, on the other hand, found the whole affair mesmerising.

In our various comments on the subject, we warned against two extremes:

  1. Overbearing regulations and omnipotent regulators whose arbitrary, mission-creeping, and cookie-cutter approach to their work can stifle innovation and prevent genuinely mould-breaking innovations and innovators from entering our financial markets;


  1. Weak regulation that fails to engage with emerging trends in the financial markets, favour strong competition, or promote fair and transparent standards; and the absence of a consumer protection regime premised on open and consistent communications, clear guidance and risk-based supervision.

The first warning had the Bank of Ghana (BOG) in mind. The second was implied criticism directed at the Security & Exchanges Communication (SEC).

BOG was at the time of our initial commentary fast transforming itself into an octopus-omnibus regulator and encroaching the territory of other regulators; whilst SEC seemed, to all intents and purposes, comatose.

Since then, a whole lot has happened on various fronts, but the tension brought about by the continued framing of the Menzgold affair in certain circles, even in some sophisticated ones, as a tension between “financial innovation”, on one hand, and “firm regulation”, on the other, continues to block effective analysis. One Academic, an economist, was recently heard on primetime television applying the precautionary principle of harm: if people are benefiting from Menzgold, then what value would regulation bring, he asked.

In many ways, that is the issue at stake: innovations bring benefits to many people, and unless regulation can show that risks or harm to other people are possible, or even imminent, in the absence of control, many liberal-minded people are unlikely to accept that regulation must by all means be imposed. This country does not exist to provide regulators with “something to do”. Regulators, like all other professionals, must continually convince us that their resources and powers, all granted by public endowment, actually do improve our society and ultimately enrich our lives by securing our interests.

Menzgold insists that they are financial innovators here to rescue us from a moribund financial system that is shortchanging consumers by providing returns below inflation, and that their presence implies much needed competition to mutual funds, unit trusts, and other investment platforms in the country, who have had it too easy for far too long.

The regulatory agencies insist that, even so, Menzgold needs to comply with the relevant laws so that their operations can be better monitored to confirm that they are indeed doing what they claim is in conformity with the law.

Unless it can be shown that the activities of Menzgold are genuinely so revolutionary that our laws as they exist today did not foresee them and therefore could not have determined that they have characteristics that warrant additional regulation beyond that which all businesses must comply with (eg. business registration and tax), Menzgold would need to undergo immediate licensing. The demand for a “special” dispensation to accommodate them would be shown to be empty.

Licensing would ensure that they file regular reports to the authorities and submit themselves to intrusive scrutiny including the verification of claims made in the reports. A major feature of such prudential regulation would be to establish that the company is technically solvent, and that it is observing the standard rules of accounting practice.

Sections 24 to 31 of the primary legislation on securities in Ghana (Act 929) grant wide powers to the SEC over all businesses in Ghana dealing in securities, in whatever form. This extends to authority to cause the production of accounting records and other “books” of the company under scrutiny for examination. Indeed, the SEC’s powers here may even extend to non-licensees (under subsection 24(1)g and elsewhere).

The usual argument one hears in this connection is that Menzgold has not been known to default on its obligations, so what is the problem of the regulators? Well, by that logic, any driver that has never had an accident shouldn’t worry about licensing. Licensing of drivers exist because we consider driving to be risky and potentially harmful enough that a particular type of review of drivers’ skills and vehicular conditions has been established by law. This is without respect to any particular driver’s track record.

The rational question to ask therefore is whether the activities Menzgold are engaged in currently bear similar risks to those being engaged in by insurers, brokers, banks, discount houses, money transfer companies, etc. If so, then the next question to ask is whether, beyond the similarity of risks, the activity is sufficiently similar or perfectly identical to any of the explicitly identified lines of businesses regulated by mandated agencies under the law. If the answer is yes, then the analysis moves on to determine the appropriate manner of bringing Menzgold into compliance with the appropriate licensing regime.

The simplest way to make that determination is to simply examine the “product” and/or “services” that Menzgold is selling. The most definitive description of that product/service is to be found in the legally binding contracts that cover or governs the sale. They describe, more than anything else, what is officially on offer.

We have done just that, and our findings are as below.

  1. In each of the trading contracts Menzgold enters with counterparties, it identifies itself as a “Gold Trading Platform” and the counterparty as a “trader”. Menzgold starts by projecting itself as offering an over-the-counter channel to certain trading exchanges. The “trader” utilises Menzgold’s proprietary platform to “trade” in gold in these unnamed gold exchanges or markets.


  1. Menzgold asserts as follows: “the company, under full corporate authority and responsibility, declares that he (sic) has the clear and qualified right to deal in gold (AU metals)”.


  1. In the same clause 1, the “trader” (i.e. the individual or institution entering into the contract with Menzgold) affirms that (on the pain of “perjury”, no less) they did not get the gold from a criminal source.


  1. The trader is then provided with specifications that the gold must meet:  A. The gold must be Aurum Utallum; B. It must be 22+ carats; C. It must be in the form of Gold Dore bars; D. It must have 92% minimum purity.


  1. There is a “returns price payable” that will “not exceed” 10% “extra value on the prevailing market price at the time of commencing gold trade”. The clause that follows then talks about a monthly payment schedule but does not specify the rate.


  1. Menzgold commits to “superintend the gold trading within the trading period authorised by the trader”. Note the abrupt shift from offering a “gold trading platform” to offering an “agency service” to trade gold on behalf of the counterparty/customer.


  1. If the “trader”/customer/subscriber/counterparty is also not interested in monthly returns or the 10% “extra payable value”, then by giving Menzgold seven days’ notice and paying an unspecified fee, they can come for their gold upon the expiration of the term. Considering however that the minimum term is 6 months, and a monthly schedule for returns is provided, even if the rate is not specified, how exactly one can opt out of the benefit, or would want to, remains a mystery.


  1. The parties to the “trading contract” consent to a third-party assayer determining the quantity and quality of the gold, presumably in the event of a dispute. The procedure for appointing this assayer is not specified. Our interviewees confirm that they certainly did not sight any assay report at the start of the relationship.


  1. In clause 6 of the contract, the trader/customer consents to pay a commission of 100 GHS for each 7.7 grams of gold (200 GHS for every 5800 GHS) traded on the “gold trading platform”, but this amount or quantity (it is unclear which) is “subject to plus or minus 20”. Like many provisions in the agreement, it is impossible to make head or tail of this.


  1. The entire agreement is held to be subject to certain International Chamber of Commerce (ICC) non-circumvention rules promulgated in Paris, France, at an uncertain date. The intention is perhaps to reference one of the various ICC model contracts for intermediaries (such as pub. 169, which offers boilerplate language on non-circumvention). The agreement ends with another opportunity for parties’ avowal of authority on the pain of “perjury”.

The first thing that strikes the reader is the sheer improvisational tint of the whole 4-page contract. It clearly was not crafted by anyone who has ever studied any law, much less an actually practising lawyer. The terms are contradictory and the agreement’s worth as a reference document in the event of a dispute is zero. The two main pages of the document are attached for reader’s own assessment.


Here are some of the most spectacular species of weirdness:

  1. In the particular contract we analysed, which we have confirmed to be the company’s standard template, the amount of money involved was $230,000 or 1.15 million Ghana Cedis. From our preliminary analysis based on interviews conducted so far, we estimate the aggregate value of “funds under management” or “trading volume” in the so-called Gold Vault trading platform controlled by Menzgold at $300 million. But this is based on pure extrapolation from limited data and could be off by a considerable margin below or above. Yet, the full terms and conditions of the only subsisting legal agreement used for these transactions come to less than 600 words. A standard, plain vanilla, savings account at Citi, Barclays, or any serious Bank, comes with an agreement adorned with 10,000 words. The average savings balance in the UK is $5500. We must thus add “legal innovation” to the list of innovations we are examining in the Menzgold repertoire.


  1. A “dore bar” is an alloy containing gold and lesser valued minerals that is usually the starting point of the refinery process. To specify a purity of 92% or 22 carats minimum whilst insisting on “gold dore” bars is to indulge in meaningless specification. It shows a complete disinterest in traditional gold trading.


  1. The use of the phrase “Aurum Utallum” is a tell-tale sign of a lack of professional exposure to international gold trading, as that term is meaningless Latin, and is never used in any serious international commodities trading context.


  1. The agreement effectively asks the customer to bring unrefined gold from mines to trade on a platform, placing the burden of ensuring that the gold is of the requisite specs on the same customer. This is seriously comical seeing as the customer in fact brings money to purchase the gold from an affiliate of Menzgold, and rightly so since only licensed traders can buy raw gold in Ghana anyway. The contract is effectively describing a process that is contrary to the actual practice.


  1. The wording of the “consideration” in the agreement is incoherent and incongruous. The provision that the “returns price payable will not exceed 10% extra value on the prevailing market price at the time of commencing gold trade” is at best indiscernible and at worst deliberately obfuscatory.


  1. Even the quantity of deposits is left subject to indeterminate third-party verification.

In short, the agreement is so loosely written, to put it mildly, that as a private contract, it is well-nigh unenforceable. Is this “financial innovation”?

Menzgold insists that their model is “commodities trading”, a specie of financial activity that is simply, according to newly hired, highly expensive, lawyers, unaddressed by our laws. We can play a mind game with commodities trading that could perhaps, even if murkily, enable a model to fall through cracks in the current regulatory architecture. So let’s try.

For example, if people indeed bought refined gold from goldsmiths, and sent same to Menzgold, which then “borrowed” the gold, sold it abroad, repatriated the proceeds, kept a fee, and paid the owners of the gold, Menzgold would escape the PMMC regime but it would still be caught by the “investment advisory” and “brokerage” elements of the securities legislation (Section 3(c) of the Act). But if indeed, it merely arranges for the refining of the gold overseas, deposit of same in bullion vaults, authorising of trading by an overseas agent, and repatriation of the proceeds for management fees, then the situation becomes complex.

The following is the chain of actions inherent in the hypothetical model above. Raw gold is, say, provided by licensed gold buyers in Ghana. The gold is sent overseas for refining. The so-called Gold Vault platform enables the purchase of refined gold by the public (standard e-commerce marketplace). The refined gold is thus purchased by the retail investor using Menzgold’s electronic platform locally and in local currency. The purchased refined gold is deposited overseas in a bullion depositary. A nominated agent exploits spreads in the futures market or some other arbitrage-seeming opportunity. Gains are repatriated to Ghana and paid to the buyers of the refined gold. Menzgold keeps a fee.

Would this be innovative? To the extent that no such product exists in Ghana today that leverages the full continuum of the gold trading opportunity, one cannot begrudge its innovative character. Though we know for a fact that under current market conditions this model would most likely be unprofitable, we cannot presume to know all futures trading algorithms in the world available to every possible agent that Menzgold can engage.

Would the said business model escape the ironclad jaws of the SEC though? No. There is still an element where Menzgold is serving as a “clearing and settlements platform”, and to the extent that money is paid at one point to an institution, Menzgold, and returns received later from the same institution, in the expectation of profit, by an investor, one cannot escape the designation of “investment security”, within the general meaning of that term. In simple terms, even if the high bar of “innovation” could be met, that would not automatically absolve of the pain of regulation.

The fact that the Act does not provide an all-encompassing definition of “commodities futures, contracts [and] options”, or of “settlement” affords no comfort for an evader whatsoever. In the same way that “derivatives” are not defined in such a manner as to explicitly adumbrate all the million and one varieties of derivatives seen in many markets around the world, there is a certain sense in which the broad definitions are purposefully broad to capture reasonable varieties of actions obvious to those skilled in a particular industry.

The definition of “derivatives” in the Act, as a “financial instrument” whose value is derived from one or more underlying “assets” is so on-point as to make the argument that the Menzgold offering does not constitute either a derivative, or at the very least the means to trade derivatives and options, completely untenable. In a court of law, the exact mechanics of how exactly Menzgold converts notional credits of gold allocated to people’s account into profits in a world where the price of gold has been falling would be laid bare without the protections afforded purported trade secrets, which is probably why Menzgold has wisely chosen not to approach the courts.

So, we return to the question: is Menzgold a financial innovator?

The answer is clear: not according to the product description in the contract covering the products they are selling. The design of the contract shows striking inattention to detail. If anything at all, the quality of the contract raises great doubts about whether Menzgold understands the financial industry even well enough to actually participate in it.

Does that mean that there is no risk of regulatory overreach? We reckon that there remains such a risk. We should all not be overly fixated on Menzgold when evaluating the general posture of the regulator. We can be as concerned about the operational behaviour of regulation as we can about its substance. We have to be wary of regulatory overreach because otherwise someday regulators could stifle real innovation if we let down our guard.

There is also the issue of “political economy”. A regulator has to be sensitive to its environment. We live in a society where attention to the fine details of situations is not the forte of even the elite, not to talk of the masses.

A forced and abrupt winding down of Menzgold is very likely to be extremely nasty, politically and socially. Given the opacity of the company’s operations, and the fact that it has been unregulated, and therefore never reported, for nearly five years, the first order of business should have been intelligence gathering and the obtainment of detailed information about assets, including assets held overseas. In particular, an audit of Menzgold’s bullion position is extremely critical. It is the first stage in determining whether it has the capacity to meet even the bare obligation to return the 75% of the gold it is purportedly trading on behalf of their owners back to them in the event of a liquidation (the agreement it signs with its customers does not detail what constitutes a “force majeure” – a contract-relieving incident).

Section 205 of the SEC’s Act anticipates situations where the Commission would encounter recalcitrant companies. It thus makes provision for the use of court orders to compel undertakings. SEC could have used these powers to secure as much access to the information held by Menzgold so that prior to an order of suspension of trading it would have been in a much better position to determine the company’s risk of default on the hundreds of millions of Ghana Cedis’ worth of contracts it has entered into.  The approach that the SEC, and prior to it, the BOG in particular, seem to prefer may lead to a situation where the regulators are blamed for any defaults that occur.

We acknowledge an argument, gaining currency, that people who can “deposit” notional gold worth more than a million Ghana Cedis on the strength of a 4-page agreement ridden with elementary legal and technical errors and mindboggling confusions deserve to lose their principal, and not just 75% of it.

Unfortunately, from what we can gather, financial institutions and other types of institutions are considerably exposed to Menzgold.  There have been substantial placements as well as roundabout placements, of investments, through borrow-to-play methods, with the self-declared “gold dealership”. Should a mass default be triggered without proper due diligence, the ripple effects could well bring down another group of financial institutions, especially in the lower tiers of the banking system, which are already under great strain.

In short, we need savvy regulators fully capable of innovative risk management. Who said only entrepreneurs need innovation?

Selorm Branttie and Bright Simons are affiliated with IMANI Center for Policy & Education, which does not, as yet, endorse these views.


There are reports in the Ghanaian press that the IMF has referred Ghana’s recently ratified Master Project Support Agreement with Sinohydro Corporation of China to its legal department for advice on whether what the Ghanaian government is calling a “barter agreement” should instead be classified as a loan.

Intrusive as that may sound, and an affront to Ghana’s sovereignty as some may rightly term the intrusion, it may well serve some use, considering the fact that the said agreement has not been made accessible on the website of the Ghanaian Parliament or any government website for that matter. Most analysts may just have to wait for the IMF’s legal opinion.

Luckily, for some of us, it is not the legalities that are of primary interest; it is, rather, the commercial and financial logic.

The “bauxite barter deal” between Ghana and Sinohydro is not actually very different from the cocoa deal that got the Bui dam built (an interesting point considering the current dispute over the eligibility of the “loan” categorisation). The same company, Sinohydro, by most measures the world’s largest dam construction company, is involved. Clearly, it is drawing on its experience in the Bui episode as it enters into this latest arrangement.

Whilst the general design of the two deals, resource-backed infrastructure financing, is the same, there is a world of difference, however, when one takes into account the level of complexity.

In the case of Bui, Sinohydro was operating in its sweet spot: dam construction. A single dam to be precise. Dams are revenue generating objects, and to underline that fact, an escrow account was set up so that proceeds from power sold to the Electricity Corporation of Ghana (ECG) could be channelled directly to offset part of the debt. Furthermore, the natural resource involved – cocoa – is a mature category in Ghana’s portfolio of assets. The buyer – Genertec – and the ultimate lender – China EXIM bank – could be assured of the volumes without much fuss (about 40,000 metric tonnes per annum, just about 5% of Ghana’s usual annual production).

The “bauxite barter” deal on the other hand is far more complicated.

In the first place, the actual resource being presented as security is not yet available. The Government of Ghana chose “alumina” instead of bauxite in order to improve the value of the security. The motivation is very easy to understand, at an 8% discount rate, Ghana would need to find bauxite worth $420 million a year over the 15-year life of the facility for the arrangement to make sense (assuming zero production costs, a subject we treat later). With Chinese landed bauxite prices hovering around the $53 mark for our type of bauxite, that would suggest about 8 million metric tonnes of bauxite a year. Ghana has been struggling to produce even 800,000 tonnes a year to date. The infrastructural investments required to triple this quantity by bringing the Atewa reserves into play are significant (assuming here that all bauxite produced into the country shall be handed over to the Chinese). Increasing the overall production by ten times, which is what would be needed were bauxite the resource collateral, is simply unrealistic for at least 8 more years.

Hence the focus on alumina. Sound pricing forecasts for alumina in the near-term indicate an average of about $550 per metric tonne. In essence, using a rough rule of thumb, 1 million tonnes of alumina requires just about 2.5 million tonnes of bauxite, in turn requiring that Ghana only triples its current production of bauxite (assuming, once again, that no bauxite shall be exported raw to any other country).

Tripling bauxite production should not be gut-wrenchingly difficult. All that is needed really are access roads, giant trucks, ore cleaning machines, and port storage facilities. At those volumes, rail investment can be deferred. We can be safe in the knowledge that with an estimate of $100 million in capital expenditure and $25 per tonne in operating costs we would be getting close to what is required to deliver our side of the bargain. That means that an upfront investment of about $125 million should be sufficient to unlock the $2 billion. Except of course that it won’t. We are not carefully accounting for the alumina refining component of the deal nor explaining who bears the operating costs for mining the bauxite.

If Sinohydro is to become responsible for the operating costs of mining the extra 1.8 to 2 million tonnes of bauxite we need to produce in Ghana in order to be able to refine the quantities of alumina needed to service the $2 billion facility, then it needs to find roughly $750 million to underwrite that expense (over a 15-year period). It seems obvious that this amount needs to be factored into the calculations somehow. That cost, carefully reviewed, is the cost of the bauxite that has to go into the alumina production.

Ghana’s bauxite is mainly gibbsite which, compared to trihydrates, consume significantly higher amounts of power during the process of conversion to alumina. That fact, compounded with scale factors and other complications, yields a production cost for alumina in the Ghana context of about $300 per tonne.

Thus whilst Ghana could easily make $550 million per annum producing 1 million tonnes of alumina after tripling its bauxite production, it needs to find about $1 billion in upfront investments, and account for running costs of about $300 million a year.

Two questions therefore arise: is Sinohydro investing this amount of one billion dollars separately from the $2 billion “barter” arrangement, or is this investment requirement factored into the $2 billion “barter deal”? If so, how come Ghana has tendered a laundry list of roads and hospitals worth $2 billion, out of which an initial tranche of $500 million is already due by the end of the year?

More critically how is Ghana going to fund the refinery costs of $300 million a year if all the inbound $2 billion is expended on roads and hospitals and no portion of the amount is amortised to assist with costs of production and maintenance? Recall that at an 8% discount rate, the average servicing costs for both principal and interest is about $420 million per annum (including provisioning for principal retirement). Should the sale of alumina yield $550 million per annum, the surplus recorded is only $130 million, not enough to keep the engines running. The situation is only saved if $170 million is available annually to cushion the financial model.  To keep matters simple this analysis does not even take into account the 15% co-financing obligation of the government, another $300 million millstone.

But even if we have been overly stringent in our projections, it seems very likely that a substantial proportion of the $2 billion being borrowed is required just to support the underlying venture and thus uphold the security of the facility (i.e. to mine and refine the bauxite). It is hard to see how anything less than half of the amount will do. Unless Sinohydro is in the business of losing money, why would it guarantee $2 billion of loans from China EXIM Bank or other lenders for Ghana without factoring into the financial model somehow the costs of securing the natural resources needed to pay for the interest and principal? Unless the profit margins on the infrastructure it is to build are so large that it intends that ultimately Ghana gets far less than $1 billion of infrastructure for $2 billion worth of bauxite. This is after all, one giant sole-sourcing arrangement.

This discussion has centred only on the $2 billion package. As everyone now knows, the government plans to secure $10 billion overall. The analysis breaks down completely in the face of such audacity.

Much of the thinking propelling these bold borrowing plans is based on a misguided notion of zero-cost production. Firstly, there was a claim that Ghana has bauxite reserves valued at $460 billion, by none other than the Senior Minister himself. This implies 8.7 billion tonnes of proven bauxite reserves, making Ghana the owner of the world’s largest reserves, and with more than 25% of the global total. Obviously ridiculous.

Even the uncritical 960 million metric tonnes of bauxite reserves figure one hears often in government circles should translate to $50 billion in terms of historical average prices. If the $460 billion figure emanates from an assumption that all that bauxite shall be converted into alumina, then one wonders why we don’t assume that the bauxite shall be converted to aluminium, or even roofing sheets.  The truth of course is that converting any part of that bauxite into alumina, aluminium or roofing sheets requires billions of dollars of investment, a prospect far from assured.

It is also instructive to point out that the 960 million tonnes of reserves number is based on wild guessing. Successive Ghanaian governments have to date refused or neglected to conduct any serious mineralogical surveys, with the sad result that most of the data used for estimation in these matters date back to the ‘70s and ‘80s and rely on half-baked datasets. The upper bound found in the only major Ghanaian geological survey of bauxite reserves to date was 580 million metric tonnes across thirteen main deposits. These were however not all proven. Proven reserves were in the order of about 370 million metric tonnes. This would suggest a valuation of about $20 billion.

One may be tempted to argue that resources worth $20 billion are still far more than needed to secure a $10 billion facility, after all it would suggest that Ghana is “leveraging” only 50% of one asset category for massive development. Unfortunately, it is not that simple. “Exploiting” reserves of any mineral, as we have seen in the preceding discussion, is a complex undertaking. Guinea for instance only produces about 0.5% of its total reserves per annum. Australia produces about 1.25%. Jamaica produces about 0.4%; and Vietnam, about 0.05%. If Ghana is successful in exploiting an incredibly high proportion of 5% (i.e. decide to consume all the bauxite it has in 20 years), it will immediately become the 6th largest producer in the world, overtaking Russia, Jamaica and Vietnam, all countries with reserves many many times the quantity of Ghana’s. This amount of production – 18.5 million tonnes a year – will still yield only about $500 million unless the country succeeds in converting most or all into alumina. Doing so will indeed yield about $4.6 billion per annum (a testament to the incredible shift of the bauxite-alumina value ratio from 6.5 to nearly 10 over the last decade). More than enough to securitise a $10 billion facility with a 15-year tenor. But it will require anything between $25 billion and $30 billion in upfront investments. And a considerable amount of environmental damage. None of which we can afford for at least a decade.

In short, the idea of using bauxite resources to secure $10 billion is fanciful, and will go nowhere. The $2 billion package is more realistic, but the undertaking needs considerably more work to ensure successful realisation within the term of the current Ghanaian Administration, as well as value for money.



1. The Government of Ghana has propositioned the Chinese for $2 billion to set up an integrated bauxite to alumina value chain in Ghana.

2. Let us assume that if they get this money they shall invest $1.4 billion through Sinohydro, a Chinese contractor, to develop rail, refining and related infrastructure, thereby adding about 2.5 million tonnes of gibbsitic bauxite production and refining capacity. Let’s assume they shall keep $600 million as working capital.

3. 2.5 Million tons of bauxite = 900,000 tonnes of alumina. Note: a 900,000 ton alumina refinery could cost $800 million on its own, leaving $600 million for investment into the actual mining operations and transport infrastructure (rail lines, rolling stock and/or access roads and mega-trucks etc.)

4. Production cost of alumina in the hypothesized value chain = $320 million.

5. Realised price of alumina (three-year historic average) – $550. Realised revenues for two years: $1B. Cost breakdown: $640m (production), $110m (GSA), Loan Servicing (at 8% nominal) x 2yrs ($320m). Net margin: -$70m.

6. These ballpark calculations strongly suggest that below an alumina price of $600, the integrated project has a high risk of unprofitability. For most of 2017, the price of alumina on the world market hovered under $350 before a massive surge to $700. It broke the $800 a few months ago but the stable trend has been in the mid $400s. This analysis is therefore considerably optimistic.

7. Creative ideas are seriously needed to improve the viability of the project.


Many Ghanaians may not be aware, but their country is actually going through a second “decolonisation” phase right now.

The “neo-colonialism” that the country’s first President, Kwame Nkrumah, railed against so presciently is actually in fast retreat. At least as far as policy decision-making in Ghana is concerned. A few more sentences and my point would be clearer.

In 1998, the Ghanaian State collected about $1.2 billion in taxes. Foreign governments through their aid agencies – such as DFID, GTZ, USAID, DANIDA, JICA etc. – handed over to the government about $240 million in grants. Of the total planned expenditure of $2.13 billion, $600 million was financed by aid agencies through grants and concessional (low interest) loans.

Fast forward 20 years to today,  the phrase “concessional borrowing” has almost vanished from the country’s policy dictionary. Apart from a $191 million receipt from the IMF this quarter, low-interest loans from aid agencies have stopped being a thing in Ghana. Total grants from aid agencies and foreign governments shall amount to no more than $128 million this year. The Ghanaian authorities shall, on the other hand, be collecting taxes of nearly $8 billion.

In less than one generation, the West African nation has moved from European, Japanese and American governments financing more than 30% (40 to 50% in the early 90s) of its budget through handouts and low-interest loans to them financing just 3% (with the handout parts making up just 0.2% of GDP today compared with 8% in 1998, a 40x drop).

Whilst it is true that some of Africa’s good economic performers, such as Rwanda, where grants and low-interest loans still account for more than 30% of the national budget, are not yet in this state, the trend towards low aid dependence is actually strong across all of Africa’s reasonably well-performing economies. For instance, Zambia also, like Ghana, managed to drop the ratio of grants to GDP to 0.2% in 2016 before it saw an uptick to 0.7%.  Nigeria has for a long time had net aid levels of less than 1% of GDP.

What is the chief consequence of this development? Firstly, whereas a DFID Chief could get a same day appointment with the President of Ghana in 2000, today he/she would struggle to get a same day appointment with a Deputy Minister.

Aid agencies have stopped telling Africa’s major growing economies how to design their major systems of governance. For instance, when Ghana started large-scale public sector management reforms in 1994, and embarked on projects such as MTEF, PURFMAP, IPPD etc., virtually every detail had to be signed off in London, Washington and Amsterdam or Copenhagen.

Two decades later, when the country’s massive youth employment program – GYEEDA – was in full swing, officials hardheartedly approached the World Bank for support. The “consultant” they brought on board went online, copied some templates, and dumped it at GYEEDA, collected his money and walked away into the sunset. Truth is, Ghana had hundreds of millions of dollars of taxes in its kitty and didn’t really care for the World Bank’s “plenty grammar”.

There is no doubt that the incidents of grand corruption have increased in frequency and scale since Ghana started this second liberation from colonisation (or neocolonisation), marked by the symbolic “expulsion” of the IMF in 2006 (their current role in Ghanaian affairs is very different from before).

Of course, no way would a Western Aid Agency have allowed the amounts of money paid out to the likes of Woyome (a self-proclaimed “financial engineer” who was paid $32 million by the Ghanaian State for raising a phantom billion euro facility from Austria to build sports facilities in 2008) and the GYEEDA contractors (on whom the Ghanaian State lavished nearly a billion dollars, according to official reports, for all manner of harebrained schemes to “create” jobs for young people)  had money from these hand-wringing agencies been involved in a significant way.

As Professor Prempeh, a longtime observer of African policymaking, ponders over the prospect of kicking out Western policy imperialists only to replace them with “vulture capitalists” and “loan sharks” (Ghana’s high-interest loan borrowing in the commercial markets has jacked up, taking foreign debt as a share of GDP from 11% in 2002 to 55% today). Kobina Aidoo, another longtime observer of policy trends in Africa, marvels at the continued reliance on Aid for critical welfare services, even as newly tax-rich African countries happily splash on prestige projects. For example, Zambia relies on external funding for 60% of its health budget. Yet, the country is building another international airport at nearly $400 million.

All this generates complex feelings. It is impressive that Ghana, Zambia, Kenya and others have thrown off the shackles of the West’s policy colonisation. As it was when they threw off the shackles of political colonisation in the 1960s. Both achievements provide these African countries with moments of great pride and self-confidence. But it seems that Nkrumah was, once again, prescient when he said that the right to be independent also includes the right for a people to make their own mistakes and learn from them.

The only question is whether the corresponding responsibility for the consequences of an African country not learning from its mistakes are currently equitably shared between its elites, who have replaced the colonial policy overlords, and the masses, who under both the colonial and post-colonial systems are the worst sufferers of poor policy and corruption.

In December 2015, a Norwegian newspaper published an article accusing the Ghanaian government of having signed a $510 million deal with a company called Ameri, whose CEO was a fugitive from justice in Norway.

The newspaper had solicited expert opinion on the deal and been categorically told that the transaction amount had been bloated to the tune of $290 million.

Whilst the Ameri deal had been noted in Ghanaian policy circles, few people had given it a second look. Ghana’s independent “policy observers” community is a tiny and highly under-resourced one. For every weird action taken by politicians that attract their attention, at least five hundred go unnoticed. Most of the activists and commentators in the organised civil society movement hold more than two jobs, and many do this work on a purely voluntary basis. So, the Ameri deal would most certainly have generated no controversy had the company not hired a shady CEO with ties to the Pakistani Mafia.

The report catapulted the Ameri matter to the front tray of IMANI’s crowded desk, and I personally took a lot of interest in it. I volunteer time as a strategic advisor at IMANI, and was previously a full-time executive.

Upon further scrutiny, it emerged that the deal was a Build-Operate-Transfer arrangement, so the $220 million price ticket of the power plant had to take into account the cost of financing. But this cost was certainly not $290 million!

Further probing showed that Ghana had been in discussions with APR energy, a respected key distributor for the manufacturers of the power plant, General Electric (GE). Indeed, all the technical documentation for the Ameri deal still bore the insignia of APR.

The question was why did Ghana start with APR, which has done several leasing and hire-purchase transactions around the world, and then suddenly decided that it needed a strange, obscure, middle man? Especially when similar deals driven directly by APR and GE were hundreds of millions of dollars cheaper than what Ameri was offering?

The argument that Ghana had no money to buy the plant upfront wasn’t sensible since the country was committing to pay more than $8 million a month and was, furthermore, also issuing, as collateral, a fat bank guarantee of $51 million (for a plant worth less than $220 million). Both the manufacturer and their principal distributors would have done a deal at a heartbeat.

Eventually the Ghanaian government opted for a deal where GE handed the power plant to APR, which in turn passed it over to Metka, a Greek company, which in turn assigned the job to PPR, its Turkish subsidiary, structured as a special purpose vehicle conveniently domiciled outside the EU.

This circuitous arrangement now in place, Ameri could sign a brokerage agreement with Ghana, immediately assign the contract to PPR, and pocket a cool $150 million (Metka had done all the engineering, and arranged all the financing to sort out APR).

IMANI’s position was that this was a preposterous arrangement and, further, that the government should do everything to claw back a good chunk of the $150 million. IMANI reviewed large amounts of data in technical reference databases and spoke to experts to double-check the Bill of Quantities provided in the Ameri contract and found massive inflation, obviously done to hide the $150 million brokerage fee.

It was obvious that Ghana didn’t have to pay any brokerage fee as the country was putting down more than $100 million a year, a fat bank guarantee, and was willing to buy its own fuel and mineral water, and provide free siting, for the power plant. GE or APR were completely willing to deal directly with the Ghanaian government, and had indeed been in detailed discussions in the period preceding the Ameri contract. Consequently, in IMANI’s view, the $150 million sweetheart agency fee needed a haircut.

Why did IMANI not recommend an abrogation of the contract? Simple: the agreement had been drafted with some of the most ironclad clauses known to man. Every eventuality had been foreseen. No way could the contract be cancelled without Ghana suffering massive penalties.

When the new Government came into office, it commissioned an expert report which concluded in the same manner that IMANI had: the agency fee was unjustifiable. The committee advised that the contract be terminated on the basis of fraud. I took to Facebook to point out that where international arbitration was the agreed dispute resolution process, even fraud is not enough to vitiate a contract without resorting to the arbitration clause in the agreement.

On Facebook, I campaigned for the government to withhold the monthly payments as a way to force Ameri to come back to the table for renegotiation.

We are aware that monthly payments were indeed suspended for a time as the Government sought legal advice from both the country’s Attorney General (AG) and its UK based solicitors regarding the legality of this tactic.

This week, the Attorney General’s opinion found its way into the Ghanaian press and then promptly, and seriously, misrepresented by hacks of the politicians that sponsored the deal. This post is to set the record straight.

It is important to note that separate from this opinion, the Government of Ghana has also solicited a second opinion on the matter. I have it on confidence that Ms. Vicki Bright, a Ghanaian Barrister, also issued an opinion. Unfortunately, I came across this document overseas in another context and I am not at liberty to share the content here, except in very broad terms.

Regarding the AG’s opinion, here is the low-down:

1. The Attorney General of Ghana has merely reinforced what we have all known from the beginning: the Ameri contract is NOT amenable to abrogation without massive financial damage. It is watertight. It was written to survive adversity. By persons very skilled at that craft.

2. Since the agreement only allows monthly payments to be withheld if Ameri’s agents (Metka/PPR) fail to deliver the power plant and keep it running with our fuel and facilities, there are no grounds to suspend monthly payments. Whether the price has been bloated or not, the plant is working. Full stop.

3. Whilst the Government of the day failed to follow due process when entering into the sole-sourcing arrangement for the delivery of the power plant, the procurement law has a loophole to cover this: the government can assert a national crisis and use that excuse. The electricity crisis then raging in Ghana was thus the perfect cover. Whether or not the emergency plant still took more than a year to show up, which indeed it did, is immaterial. At the time of the sole sourcing there were rolling blackouts across the country.

4. Furthermore, after having entered into the sole-sourcing contract, the Ministry wrote to the Procurement Authority, which pliantly, as institutions in Ghana are wont to do, blessed the prior impropriety.

5. Ameri had very cleverly hired a law firm in very fine standing with the Government of the day, Messrs Kwame Akuffo & Co. The well-known Principal of this law firm sits on the board of critical state institutions charged with the regulation of the downstream energy sector. In fact, the President of the Republic, the AG, the Chief Justice, and many prominent ruling elite grandees are alumni of an affiliate firm, Akufo-Addo, Prempeh & Co, where this Principal has practiced for many years. The firm is versed in regulatory matters. This law firm had already written to the Attorney General to threaten all manner of penalties against Ghana should payments continue to be withheld.

6. The AG’s legal opinion then goes further to lament what we all know about the agreement. Even after payments had commenced after legal close, the $51 million standby letter of credit (“bank guarantee”) was crazily allowed to remain operative. The absence of a proper power purchase agreement overwhelmingly tilts all the advantages to Ameri’s side.

7. Whilst the agreement signed is obviously very unfavourable to Government of Ghana, merely saying it is unconscionable does not provide a basis to wiggle out of it. In the eyes of the law, it is the same Ghanaian Government (notwithstanding a change of administration) that with its eyes wide open entered into the agreement. It knew it was unfavourable when it signed it. The government has never disputed any of Ameri’s invoices, and had indeed commenced some payments in 2016. It cannot all of a sudden scream “abuse!” and back out of its legal obligations.

8. Despite loud protests against the agreement, the Executive and Parliament both with their eyes wide open consented to an agreement that was clearly known to have excessive penalty clauses and exorbitant terms.

9. That the $150 million agency fee is not merited because Ameri had done no work is neither here nor there since the Government of Ghana ought to have known this fact when its agents freely, and being a government, in a superior bargaining position, consented to sign off that amount to Ameri.

10. The agreement does not allow the arbitral proceedings, should the government elect to enter arbitration, to happen in Ghana; they must take place in London. Government has the option to assert that there is fraud and corruption, but it will still have to go to arbitration subsequent to any move to terminate. Government also has the liberty to continue attempts to renegotiate the agreement.

So, as anyone with basic comprehension skills can attest, at no point in the opinion does the AG say that the agreement was sound, clean or good. It agrees that it is a one-sided and unfavourable agreement. But this was a legal opinion. It was not an opinion on the moral quality of the contract or the transaction. And as far as the law is concerned, a person, as well as a government, is free to enter into a one-sided agreement.

How the media could translate this to mean that the agreement or the transaction has been given a clean bill of health seriously baffles me. I won’t be surprised if the opinion wasn’t even read at all by the editors that signed off on those stories.

At any rate, the opinion offers nothing that wasn’t widely known already.

The decision to brave the risks of refusing to make monthly payments to Ameri, whilst still refusing Ameri access to the facility to shut it down per the penalty terms of the agreement, is a political one, not a legal one. That decision would have been very much the government’s prerogative prior to seeking legal advice, a course of action that is clearly naive. At this point it would not be wise to flagrantly ignore the government’s own legal adviser. What exactly was the Government expecting the Attorney General to say: flout the agreement and don’t pay? This is not the type of act for which one must seek legal blessing.

Going forward, the government of Ghana needs to tell its citizens the steps it is taking to prevent such poor and completely unconscionable agreements from being signed in the future. Ghanaian policy observers are now vigilant to the modus operandi:

A. Write to the Public Procurement Authority for blessing.
B. Sometimes, even secure parliamentary ratification of the transaction.
C. Put in place a watertight contract, with outsized penalties if Ghana tries to terminate.
D. Inflate your way to millions.

How does Ghana stop these kinds of situations from arising in the future?

As far as compelling Ameri to return to the negotiating table for that haircut is concerned, IMANI continues to maintain that it has always strictly been a matter of political tactics. No legal maneuver is readily available. Particularly as the government is in fact torn between two competing legal opinions, one of which I cannot discuss here.

The people of Ghana deserve a proper explanation from their Government about the status of this sad episode in their national life.

There is a very important reason why “think tank/policy activism” is completely different from academic work or academic thought leadership.

Academic work is about steadily making incremental, highly specialised, additions to the stock of knowledge, usually through peer-reviewed work. It is rigorous and tough. But it cannot respond rapidly to the fast paced policy environment.

More importantly, one cannot effectively critique any policy from just one academic specialisation. It is not practical. A multidisciplinary team can do it, but the costs will be very high, and societies like Ghana haven’t decided yet to support policy activism.

That means the only kind of policy activism that works in a place like Ghana must be driven by “well informed generalists”. In fact, because the policy space is highly contested, specialists can more easily be coopted or silenced. So one must be an activist first and a policy enthusiast second to be effective.

Another, more fascinating, reason why generalists make for more effective policy activists is the “crazy novelty” of policy problems.

Here is a tip a decade and half in this policy activism space has taught me: when an unfamiliar problem that is poorly covered in academic research, such as telecom tax assurance, pops up to confuse you, look for “equivalent domains” and take ideas from there and quickly apply them.

This also, of course, requires one to be comfortable about being an unabashed generalist. That is why in policy activism, rigour is not everything. As oldtime World Bank policy promoter, Kafu Kofi Tsikata​, puts it: ‘”rigorous enough” is fine. Speed and impact are just as critical.”‘ (Paraphrasing).

I like examples. So let me illustrate. One quick way to convince yourself that there is something messed up about how much the Ghanaian State is spending in telco tax assurance is to look at the oil and gas industry, where production monitoring does have a very important role because governments usually own a share of the oil and gas produced (i.e. direct, live, monitoring is not only for tax purposes).

Then ask yourself: how much does Ghana make from oil and gas annually?

How does the State monitor production?

What systems does the State deploy to monitor production?

At what cost?

How does that compare to what is being done in the telecom sector.

Of course, one has to be fairly familiar with the oil and gas sector to be able to run this quick comparison. But that is the whole point of policy activism (in that regard, it is remarkably similar to “management consulting”, where consultants often have to get up to speed with unfamiliar disciplinary areas within a few weeks or even less). The trick is to have a reliable and expanding rucksack of methods and analytical tools that can be rapidly used to gather critical data and extract sufficient meaning from it, adding context through pattern analysis developed throughout a career of intellectual exploration.

Let us actually try and answer the questions above so that I can advance the point.

A. Ultrasonic flowmeters and remote observation centers are often used to monitor oil and gas production, storage and distribution.

B. A good example of such solutions is ABB’s ProcessMaster (not just the flowmeter, but the vendor’s recommended end-to-end system) and Haliburton’s DecisionSpace. Lesser known systems include PetroDaq.

C. A full solution covering the scale of oil and gas fields in Ghana can be commissioned, complete with a remote mount transmitter and observation deck, for about $450,000. Since the Ghanaian authorities tend to focus only on export terminals, I reckon two sets of monitoring systems for the two FPSOs (seaborne oil and gas production and handling terminals) would cost about $200,000, but let’s be prudent and pad the costs. Let’s even say $1 million.

D. The Ghanaian State made about $550 million last year from oil revenue.

E. This means the “additional” (beyond normal GRA assurance) cost of monitoring and assuring oil and gas revenue was 18 pesewas (4 cents) for every 100 GHS ($22) collected. Now consider what the same state has been spending on Communications Services Tax monitoring so far: 24 GHS ($5.3) out of every 100 GHS ($22) collected!

As soon as you see that as a Policy activist, you start to smell blood. Of course these are two widely different industries. But it gives you a rough starting point when deciding to pursue or ignore. You see how generalism works in this context?

Regardless what perceptions one may have about generalist policy activists who seem to flit from discipline to discipline, the fact is policy activism cannot be done from a specialist standpoint. At least, not with the resources available in a place like Ghana.

Yes, you will find a lot of ignorance parading as sophistication on this matter, but just observe how academics and other specialists in Africa find it difficult to engage with policy, and draw your own conclusions.

We can have generalist policy activists or we can have nothing. Simple.


A controversy has been raging in Ghana in recent weeks: the country’s Ministry of Communications says it doesn’t trust the tax declarations of the telecom network operators and want a way to independently gauge their traffic and consequent revenue.

This is despite the fact that the telecom companies pay at least 35% of all corporate tax in Ghana, despite the share of telecommunications in GDP being less than 3%. Anyway, the government passed a law a year ago requiring this independent verification, giving legitimacy to a practice that started a decade ago.

So far, the country has been paying in excess of $30 million a year to pursue this independent verification agenda. This year the contracts of some of the companies it has hired to this work were revised and one terminated, but the net effect remains the same. Roughly $18 million goes to one contractor for monitoring call volumes and network fraud, and an unclear amount, but certainly not less than $12 million, goes to another for mandatory interconnect services, which were also justified on grounds of minimising telecom sector fraud (in most countries, telecom networks enter into direct arrangements to interconnect or decide on their own to use a clearing house.)

There is a growing concern that these programs are ineffective, and are merely enriching vendors. Everyone agrees that only about 20% of all the taxes telecom companies pay are sensitive to this kind of topline revenue monitoring, if it does in fact works. These taxes – notably the Communications Service Tax – net less than $80 million a year. To spend nearly 40 cents to monitor, assure and collect each $1 is highly irregular. Ghana generally spends about 2 cents for each dollar of corporate tax it collects. In the UK, the costs are half that amount for equivalent revenue per taxpayer.

Having some time on my hand this evening, I decided to break down the problem, and design a solution from the bottom up.

In tribute to the Godfather of “tropically tolerant tech”, Herman Chinery-Hesse, we are calling it: “The Nii Tettey Telco Tax Assurance System” (NTETAX).

The chief design constraint was that it should cost no more than $55,000 a year to set up and run, but it should be more effective than the Afriwave-Subah-KelniGVG model that has been in use since 2015 at a cost of more than $30 million per year by a significant factor. I approached the problem from a Value Operations Methodology standpoint. The process has been, at a very high level, graphically described in the figure above.

Here is how NTETAX will work.

A. A GSM SIM gateway with modem-router combo is set up with 6 racks containing 960 SIM cards. Set-Up Cost & Maintenance for 1-yr = $2200

B. An HP proliant RAID to store data, host databases, and mount application servers. Hardware and software firewalls and security filters. Setup Cost & 1-yr Maintenance = $9,000.

C. A SAAS subscription package for Equinox or Neon CDR analysis and re-rating software. Cost for 1-yr = $9600

D. Open-source operating systems and work on database setup and configuration. A small, open-source, data analysis suite (for regressions, correlations, and basic data mining). Cost = $3200

E. Personnel = $28,000

F. Physical Data Room = Gratis (NCA or NITA location – these are Ghanaian state agencies with considerable material resources)

G. Airtime and mobile money costs = $1200

Here is how it will work:

A. Randomly sequenced automatic call, data, mobile money, SMS and USSD sessions are initiated from the SIM gateway to all CST-collecting ISPs and telcos round the clock.

B. These probe-sessions are duly recorded in the database.

E. CDRs are received at the end of each working day in XML or CSV format from each of the regulated CST collectors via an FTP interface or other agreed file transfer protocol. The data dumps go into a database and are primed for analysis.

F. The Equinox or Neon software parses and generates summary statistics and conducts re-rating based on shared tariff rules.

G. Re-rated Service Charge Detail Records (SCRs) are automatically compared with telco-supplied SCRs. (“Telco” = telecom network operator).

H. The “mystery call” probes from the SIM gateway records are now meticulously crosschecked by the proprietary application with data contained in the CDRs and SCRs sent from the telcos. An error rate of 1 out of 10,000 sessions may be reasonable.

I. If the error rate is exceeded, a reaction limit module generates an alarm. This is a sure sign of an anomaly (note, however, that the mistake may not be deliberate) and a prompt for investigative action.

J. The SIM cards in the SIM racks are replaced at a constant rate.

K. Naturally, the MSISDNs of the probe SIMs have to be in an encrypted file, as breach would compromise the whole model. This system can be up and running at the NCA and NITA in 2 months. Training and handover should take a further 6 weeks. Three new personnel are required to oversee the operation in the data room – a DevOps engineer, a Revenue Analyst and Database Administrator. Existing personnel working for the Ghanaian Tax Authorities can be trained to support as investigation officers and assistant revenue analysts.

Unless some fatal flaw is detected upon closer scrutiny by some ultra-specialist, this compact solution seems on first inspection a superior replacement for the current system which relies solely on CDRs from the telcos but lacks both adequate re-rating capacity as well as a ground-truthing scheme. And of course, it costs $55,000. Even if it replaces only the traffic monitoring and anti-fraud solution currently in place, that’s still more than $17.9 million saved. Not small change in any Sub-Saharan African country.


Credit: original diagram by Aayush

Look carefully at the diagram above.

1. From 2010, and especially 2015, onwards, agents of the Ghanaian government (notably contractors, GVG, Afriwave and Subah) used to collect logs termed as Call Detail Records (CDRs) generated by the Authentication, Authorisation & Accounting (AAA) systems of the telecom network from network switches. A module of the AAA called the Service Control Point (SCP) interfaces between the live network and the software system that does the billing.

2. As you can see from the diagram, those CDR records are useless until they are “interpreted” by the charging system (the part encircled in blue). Easy way to understand: supposed you call a friend and he/she does not pick. Supposed someone just counts this call and claims that the telco collected money for it. Would that be sensible? So, in essence, the CDRs have little meaning unless the telco also provides the tariff scheme, charging rules, discount schedule (for all the promos etc), among other algorithms. That is what the part enclosed by the blue circle in the diagram does.

3. Whether due to ignorance or neglect, it has now come to light that much of what Ghana has been billed for (at least $32 million a year for a while now) so far in the name of “monitoring” lying and dishonest telcos by sending CDRs to contractors has just been driven by ignorance. The contractors have been relying on the telco’s reinterpretation of their own data.

4. Most Ghanaians appear confused by the development because they had assumed that by “monitoring” it was meant that the contractors were actually deploying equipment in the part enclosed by the green circle. That is the base station controllers. That would of course have been a joke if the point is to accurately measure revenue to the last dollar. Even the telcos themselves don’t “count calls on masts”.

5. What the Ghanaian government now says it will do is, in the words of its spokespersons, “live monitoring”. That means the telcos won’t be emailing the CDRs in batch on excel spreadsheets. The new Ghanaian contractors – KelniGVG – shall collect the CDRs one after one in a continuous stream. Unfortunately, that would still be useless without relying on the telcos’ charging software to provide the relevant context such as which calls and data downloads were charged and for how much (rating, metering and billing).

6. Government officials have thus been at pains to clarity that the contractor shall also be collecting the billing reports (service charge records). But it is the same CDRs (or IPDRs, in the case of data) that when rated become the billing reports/SCRs. And at any rate, the contractors will still be relying on the telco’s billing systems, or at least regularly updated charging schemes and rules, to convert the raw data into accounting spreadsheets that are meaningful. Frankly, it looks like some people want to hide behind jargon to hoodwink the Ghanaian people.

7. Rather than revenue assurance, what this is most likely to be useful for is customer protection. Had the Ghanaians set this up right, then this could have been used to help subscribers who believe telcos are “stealing their data” to get independent verification and perhaps relief. In most sophisticated jurisdictions, when regulators talk about this kind of stuff that’s what usually concerns them most.

8. But if at all Ghana has to continue to running its own independent CDR re-rating and charging processes (which is what this is about) there are many off the shelf packages used in the industry for just that. Telcos charge each other for interconnect services, roaming, termination, etc., all the time. Software for doing that is available on a SAAS/hosted basis from as low as $1000 a month. Even if the Ghanaian government wants to host the solution locally in its own network control room/datacenter, one struggles to see how setup and maintenance for something like this can cost more than $150,000 a year.

So why is the Ghanaian government spending $178 million on this expense for 10 years?