On 24^th September, 2020, IMANI partnered with ACEP for the first round of consultations with the public being undertaken by the CSO Alliance advocating against the listing of Agyapa on the London Stock Exchange as early as next month.

Whilst government of Ghana remains recalcitrant, the CSOs are also unrelenting in their efforts to widely draw attention to grave defects in the structuring of the deal.

Below, IMANI shall be drawing on the insights collated during the recent webinar with the public to share the ten most alarming problems with Agyapa that were identified. The webinar attracted many of Ghana’s most knowledgeable people in the resource governance, accountability and transparency sectors. 

A more intensive treatment of these issues are found in the briefing document used during our consultative forum (available here: ).

1. The government arrived at the valuation of the deal by assuming that the average annual production volume of gold across the 48 lease areas shall be 2.9 million ounces during the term of the agreement. This is woefully undervalued. A careful review of the 2019 data submitted by the Ministry of Finance to the CSOs to back this position reveals alarming omissions, missing numbers, and a poor appreciation of the fact that since 1990, gold production in Ghana has grown at an average of about 7% per year.  

More careful analysis show that gold production in the 48 lease areas is already heading towards the 4.2 million ounces mark in the near-term. Over the 21-year conservative life of the agreement (the agreement has no definite term as any renewal of the leases covering the relevant areas automatically extends the tenure), output shall average 4.9 million ounces a year using a conservative growth figure of 4% a year over the next two decades.

2. The government conveniently argues that the agreement will terminate when the last of the 48 mining leases expire. This betrays a weak understanding of the mining sector in Ghana, where extant agreements are crafted to enable continuous renewal so long as gold is being discovered and commercially mined. Anglogold’s Obuasi operations have leases that have been effectively active for over 100 years. 

The Government is thus seriously underestimating the amount of gold that can be found in the vast tracts of land that makes up these leases. Until recently, AngloGold’s lease areas exceeded 400 square kilometers. Some of the leaseholders are currently only impacting less than 10% of their concessions. As the price of gold increases, they shall aggressively a) expand development of deposits and b) start mining lower grade deposits. The government’s low “reserve life” estimates are based on a misunderstanding of how pricing helps convert mere “measured resources” into new reserves.

3. Some of the reserve estimation work is completely confused. Take AngloGold, for instance. The government limited its calculation of output in terms of  2019 reserves and duration of the output of 15 years, forgetting that it was the same government that went to Parliament in 2018 to seek considerable reliefs for the mining giant on the basis that the company shall be extending mine life by an additional 22 years, and that after 10 years the company expects to hit much higher grades of ore. The reliefs granted to Anglogold were premised on lower output today and higher production beyond 10 years. The government’s calculations are clearly muddled.

4. The Government’s use of $1300 as the average price of gold over the long-term is quite frankly reckless. Gold has moved from $393 in January 2001 to over $1900 today. Between that period, its trajectory has been mostly upwards except for a few retreats. The long-term picture of gold has thus been generally upwards. Every analyst worth its salt is predicting a long-term average above a new support of $1800. When you combine this fact with the proper treatment of the production forecasts available in the mining company’s own disclosures, you arrive at a present value calculation of more than $3 billion. The attempt to dump this valuable resource for $1 billion is unconscionable.

Read CSOs presentation on Agyapa deal

5. The government is misguided on multiple fronts when it says that “it is the market that shall value” the royalties it is giving away. There are two levels of analysis here. First, government is investing Ghana’s future gold earnings into a company. Second, it is trying to sell half of that company for cash upfront today through an IPO. There must thus be an initial valuation of how much exactly Ghana is investing before we talk about anything else. Only then can the returns: upfront cash and shares that entitles Ghana to future dividends be valued.

Ghana is investing a total of at least 5 million ounces of gold over a 21-year period (or more gold over a longer period). How much is this gold worth for how much return exactly? Government wants $500 million upfront for 2.5 million ounces of this quantity. These numbers are fixed in the agreement and have nothing to do with any “market”. The CSOs believe that by selling forward less than 1 million ounces, Ghana can get more than $500 million. The government’s current structuring means that it is selling Ghana’s gold forward at a price of $200 an ounce, when current prices are at $1900 and similar sell-forward deals have recently priced gold at $500 on average (taking time value of money into account).

It is the next stage of the analysis that involves valuing the shares in the company in which Ghana intends to invest another $100 million (as working capital) and an extra 2.5 million ounces of gold (minimum) over a 21 year period in exchange for 51% of shares. The price of these shares has already been decided by the underwriters that government has been working with for more than 2 years and set at $500 million. That is the whole point of a bookbuilding IPO. The bookrunners set a final price and offer the shares to the market. There is no planned auction for the “general market” to decide, and Government cannot make any gains on the secondary market because it must hold the 51% of shares for the long haul.

6. The government is confused about the short-term price rises that will follow the debut of the stock on the public market. These do not translate into capital gains unless government sells a big chunk of the 51% stake in a few days following the debut. Government is barred from doing this by market rules and its own policy. The people who will thus cash in are the underwriters, the investors they allocate some of the shares during bookbuilding to, and the brokers/dealers. We expect a cool $350 million could easily be made by these people due to the deliberate underpricing of the asset. After the price surge in the first few days, the stock will return to its mean. All this is on top of the massive fees they shall be earning (about 10% of total funds raise, making this method of raising upfront cash far more expensive than any debt government has raised in the last decade on the international markets).

7. Government’s expectation of 50% of retained earnings being paid out as dividends is founded on nothing. This is a company that is starting life with just $100 million of working capital, a lack of experienced management, and a non-existent operating history; and is furthermore seeking to invest in the heavily capital-intensive gold mining sector. It will need to borrow massive amounts of money at relatively high-interest rates to be competitive, and invest all the Ghana gold royalties it receives. It will have no latitude to be handing large dividends to anyone for a long time. So, future governments are being “robbed” of the highly reliable royalty income stream, which accounts for more than 90% of mineral revenues (besides oil) that government can use to support the budget. In fact, a low initial valuation means a weak starting balance sheet which more or less guarantees this outcome.

8. More worryingly, because the government also agreed to hand over effective control of the company to the minority shareholders and their appointed independent directors, it cannot dictate a dividend policy or push the company to even invest in Ghana. The government was happy to agree to this as part of the process of admission to the standard listing of the Main Market of the London Stock Exchange despite there being an option for the company to list as a “sovereign controlled commercial company”, which would have allowed government to retain significant control. 

The strategy at work here is the same one that was used to push Ghana to abandon its golden share in what became AngloGold on the basis that this will unlock more value. In 1994, before the listing of the then Ashanti Goldfields, Ghana had a 55% stake in the entity worth $880 million as share of bullion. Today, it owns 0.01% (having given up even the standard 10% it owns in all local gold companies) worth a paltry $110 million in equity value. Actual earnings come just from royalties, a measly $22 million, whereas in 1996, dividends and royalties together amounted to $115 million. Poor “internationalisation” strategies have never served Ghana’s sovereign commodity interests well.

9. The government, on top of selling gold going for $1900/oz today at $200 an ounce, then added a highly valuable option to the package. Agyapa has right of first refusal to any future royalties deal Ghana enters into. A simple rule of basic finance is that options are always valuable and must be priced. Ghana gave away this powerful bargaining chip for FREE.

10. The argument that the listed vehicle (Agyapa Royalties) had to be incorporated in the tax haven of Jersey to mirror the tax efficiency rules in the MIIF Act is untenable. Whilst returns to the SPV itself need not be taxed, investors who profit from buying shares in the SPV during the book-building period need to be KNOWN and TAXED. Even if we must go the IPO route, many jurisdictions with far better transparency standards exist, some of which have double-taxation treaties with Ghana. The same tax-efficient outcome could have been reached by using a BEPS strategy in one of these other places.

By incorporating in Jersey, where the laws permit notoriously impenetrable trusts, it has become all too easy for the underwriters to underprice the vehicle (and by implication, Ghana’s gold royalties), allot the shares to investors hiding behind trusts, and when the stock debuts, join these crony investors to pocket hundreds of millions of dollars without any of us being any wiser since such trusts would usually be “trusts of trusts”, and effectively impossible to unveil. Regardless of the government’s protestations, they have no tool to stop the underwriters from doing this. Can Ghana tolerate this risk?

Should Ghana countenance any of the above at all?