Suggestions
Bright Simons
Bright Simons is a highly accomplished individual with a strong background in social entrepreneurship and social innovation.
He is a President at mPedigree and has been recognized with prestigious awards and accolades such as the World Economic Forum designation and being named a Tech Museum Laureate.
Bright Simons is an Ashoka Fellow, showcasing his commitment to positive social change and innovation.
His expertise in social innovation has led to widespread consultation on various social issues, highlighting his thought leadership in the field.
Highlights
In 2020, Ghana had enough gold refinery projects in the pipeline to be adjudged sub-Saharan Africa's biggest gold refining hub outside South Africa.
This was of course entirely in line with its status as Africa's largest gold producer.
Why then do Ghanaians, their leaders and the general public alike, feel that "value addition" is still a distant prospect?
The answer lies in unravelling an old confusion about the link between the tangible and the intangible in "value chain transformation". Often, the intangible is lost in our analysis.
You would hear that Africa produces 70% of the world’s cocoa but receives less than 3% of chocolate sales. The focus here is entirely on the tangible: cocoa beans. Not the intangibles: branding, marketing, research & development, narrative-building, even country-image (SWISS chocolate, mon ami?)
Hence why many cocoa processing plants have struggled to make a dent in Ghana's cocoa value addition burden. In pure arithmetic terms, Ghana's cocoa processing capacity (over 500,000 tonnes per annum) has outstripped Ghana's raw cocoa output in the latest farming year (~450,000 tonnes). Yet, most processors have historically struggled with profitability. Just as most of Ghana's large gold refineries have. Some have been completely idle for years.
The challenge, of course, is that cocoa "processing" has been heavy in those categories where tangibles dominate, like powders, butter, oil etc.
True cocoa value chain transformation, however, requires a massive complementary upgrade in the intangibles aspect to produce brands and inspirations in the chocolatier's and confectioner's arts & crafts.
Same for gold. Jewellery, electric circuitry, dentistry components, aerotech - those are the contexts where gold finds its highest value expression. A nation must create value chain linkages to them, wherever they may be, not because it has gold, but because it understands them. Insight is the ultimate intangible.
Contexts full of intangible domains like R&D, "creative expression", long-term training, and "country image building". Contexts that are so hard for conventional politicians to grasp and grapple with.
Contexts which only genuine leadership across multiple segments of society can advance.
When Ghana discovered oil in 2007, some observers predicted the "Dutch Disease" as the variant of the resource curse awaiting the country.
In new research by Colin Constantine of Cambridge and Tarron Khemraj of NCF, it is shown that the classical Dutch Disease has not manifested in Ghana after nearly 15 years of oil production.
(Simplifying massively hereon.)
The classical Dutch Disease proceeds as follows
- Country discovers oil (or other major tradable natural resource);
- A sudden surge of exports generates massive forex inflows;
- The country's currency strengthens;
- Imports become cheaper, undermining some local industries;
- There is a shift of resources and emphasis to the new oil sector; and
- Legacy industries suffer and the economy comes under pressure.
Constantine and Khemraj provide evidence in their new preprint (link to full paper in first comment) that this sequence doesn't apply in the case of Ghana. It does fit the case of Angola though. As well as Trinidad and Tobago's.
At any rate, in a liberal economic setting where imports can flow without too much friction, a surge in forex inflows from resource exports should ordinarily not trigger the first step in the sequence: over-valuation of the currency.
In Ghana's case (as also in Algeria and Mexico), new oil finds have coincided with exchange rate depreciation. Ghana's situation in particular is pretty alarming. Since it became an oil producer in 2007, Ghana's currency has fallen by over 93% in value.
The Constantine & Khemraj paper spotlights one important cause of the Cedi's precipitous decline: the role of the Central Bank in the "monetisation" of the fiscal deficit.
Fiscal deficit monetisation came up in the discussions during 2022 when the Ghanaian currency lost more than 50% of its value in less than a year. However, this paper is one of the few empirical investigations into the interconnected variables at the root of the phenomenon, even if its focus was not on the acute factors that precipitated Ghana's debt default and is instead on the long-term currency movements linked to the fiscal and current accounts.
Whilst the paper does require a careful analysis of the relations among such important macroeconomic indicators as bank credit, government deposits, and resource rents, the emphasis on central bank behavior will find perfect resonance with the mood of analysts and activists in Ghana who have been sceptical of the policy conduct of the Bank of Ghana.
My personal view is that the scale and maturity of the oil industry in case study countries could also have an interesting role to play in whether monetary outcomes result in distortive appreciation or depreciation of the currency in the wake of oil finds. Still, any rigorous work that shines much needed light on the central bank's role in these times is a vital contribution to the conversation.
