^{1}

^{2}

^{3}

The Akwatia Diamond Mine has been responsible for most of the diamond-related activities in Ghana. This study analyses the factors responsible for the recent lack of investments inflows into the Mine and also assesses diamond export performance in Ghana using the annual time series data from 1987 to 2018. The Gravity Model, a well-known model for trade flow analyses is employed into the study to assess the relationship between diamond exports and foreign direct investments as well measure its impact on gross domestic product, diamond price, real effective exchange rate and inflation rate. The study found a positive but insignificant impact of foreign direct investments on diamond export over the period under study in the long run and also reports little or no direct impact on diamond exports in the short run. The study recommends that the Government of Ghana should introduce punitive laws and regulations to guide the next Akwatia Diamond Mine divestiture to avoid a repeat of the past divestiture failures. Additionally, the study recommends that the Government of Ghana should reconsider lifting the ban on artisanal diamond mining and rather formulate and adopt policies to manage and supervise their activities to boost diamond production, exports and revenue.

Total revenue from diamond exporting countries was US$127.9 billion in 2018, of which Ghana’s exports amounted to $2.1 million, representing a 0.002% of global output^{1}. This output value is an added evidence of the fast decline of the diamond sector of Ghana for the greater part of two decades. The rather unfortunate circumstance that the sector finds itself can be traced back to the closure of the Akwatia Diamond Mine in 2007 in part as a result of lack of investment and mismanagement of the mine.

The Akwatia Diamond Mine is situated in the Birim Diamond Field in the Eastern Region of Ghana ran by the erstwhile Ghana Consolidated Diamond Limited (GCDL) and has produced more than 100 million carats of diamond since it first started operations [^{2}., the governmental body who at the time was responsible for all diamond sector and diamond-related activities in Ghana.

The diamond sector of Ghana was the number one mineral produced and exported in Ghana, being the biggest contributor to foreign exchange earnings to gross domestic product (GDP) in Ghana from the 1920s until the SAPs [^{3} until 2007 when it was finally closed to operations, rendering about 800 workers redundant after the Divestiture Implementation Committee (DIC) revoked the license of Sapper and Associates, who at the time had won the te1nder to operate the Mine. Nonetheless, before the closure of the Mine, the United Nations (UN) in 2006 banned Ghana from exporting diamonds, as there was a doubt that Ghana was dealing with conflict diamonds. The Mine, however, was reopened in August, 2011 after a Ghanaian-owned company, the Great Consolidated Diamond Company Ltd. (GCDCL) won the tender to operate the Mine^{4}. In a call to government in December 2017, the Legislature of Ghana’s Parliament advocated for GCDCL’s license revocation for failing to honor its contractual obligations, six years after the said contract was signed between the two parties^{5}. According to the divestiture agreement between government and GCDCL, GCDCL promised among other things, to invest $100 million in a five-year period, with a projected production of one million carats of diamond per year within the period, creating 2500 direct and 50,000 indirect jobs.

In March 2019, DIC finally abrogated its agreement with GCDCL. The reasons assigned to the abrogation were the persistent failure of the company to settle its debts, arising from non-payment of the balance^{6} due from the divestiture of the assets of GCDL to GCDCL. This new development in the sector currently leaves the Akwatia Diamond Mine and the people of Akwatia hanging in the balance as to when and from where the next divestiture is going to come from and as to whether it would be sustainable enough to provide the necessary foundation to commence operational and commercial activities at the mine, provide jobs for the local people as well as bring an economic facelift to the Akwatia township.

The significance of this study is to point out among others; the extent to which the investments that came as a result of the SAPs and the recent divestitures benefitted the diamond sector in the short term but cast a shadow on the long term productivity sustainability of the sector. The study also seek to establish the probable profitability between diamond sector investments and diamond exports for the period under study and also the underlying factors responsible for the failure of the previous divestitures by Sapper and Associates and GCDCL. The paper also attempts to ascertain the factors that would be necessary for a successful divestiture, sustainability and profitability of the Akwatia Diamond Mine.

Capital investment is targeted at sectors and industries that promise the highest

returns [^{7} of Ghana since the introduction of the Economic Recovery Programs (ERP) in 1983 SAPs has been positioned to attract FDI inflows. Statistics show that the largest portion of FDI goes to the mining^{8} and oil industries in Africa [

· Generally, FDI has been seen to have a positive relationship with production and exports [

· The role FDI has played in shaping the economies of developing countries cannot be overstated. Thus, influencing trade patterns in China [

· FDI’s impact on economic growth and development has been widely investigated as well [

· It is no secret that depreciation of the host country’s currency might drive up future investment inflows [

· Based on varying expectations, FDI is projected to have a positive effect on GDP and GDP growth rate [

· The history of commodities exports in Africa is such that not much value is added to the exported raw materials and this affects foreign exchange earnings. Based on the statistics from the Ghana Chamber of Mines^{9} and the Kimberley Process^{10}, close to 80% of all diamonds mined are exported in their raw forms whereas the rest are processed locally to make jewelry or for other industrial purposes.

Annual time series data from 1987 to 2018 obtained from various institutions are used for the analysis in this study. The econometric software EVIEWS 10 was used for the multivariate computation and analysis.

The framework of analysis is centered on the gravity model of international trade. The gravity model measures the relationship between a trade-related response variable and selected explanatory variables such as prices, GDP, inflation, exchange rate and income [

Based on the gravity model, the basic model of export trade flows is stated as follows:

Variable | Source | Unit |
---|---|---|

1) Diamond Exports from Ghana (DExZ) | Minerals Commission of Ghana^{11}, Ghana Chamber of Mines, Kimberley Process. | Carats |

2) Foreign Direct Investment (FDI) | Minerals Commission of Ghana, Ghana Chamber of Mines. | US $ |

3) Gross Domestic Product (GDP) | World Bank | % |

4) Diamond Prices (DPx) | Minerals Commission of Ghana, Kimberly Process | US $ |

5) Real Effective Exchange Rate (REER) | World Bank | % |

6) Inflation Rate (INFRATE) | World Bank | % |

Source: Authors’ Construct.

X i j t = β 0 Y i t β 1 Y j t β 2 ∏ m = 1 M ( Z i j t m ) β m ε i j t (1)

where, X i j t is the magnitude of diamond export between Ghana and trading partners^{12}, Y i t Y j t are the revenues of the countries, ∏ m = 1 M ( Z i j t m ) β m is the set of indicators, having both positive and negative impact on trade flows. Trade flows in this instance include diamond exports (DEx), foreign direct investment (FDI), gross domestic product (GDP), internal price of diamond (DPx), real effective exchange rate (REER), and inflation rate (INFRATE). ε i j t is the random error term.

Reference [

β 0 * = β 0 Y j t β 2 (2)

The gravity model of export trade flows is therefore stated as follows:

X i j t = β 0 * Y i t β 1 ∏ m = 1 M ( Z i j t m ) β m ε i j t (3)

The general equation is usually estimated in a log linear form as:

ln X j t = β 0 * + β 1 ln Y i t + ∑ m = 1 M β m ln Z i j m + ε t (4)

The method of adding additional variables in the gravity model of international trade has been used in a number of studies [^{13}, DPx^{14}, REER^{15} and INFRATE are employed as explanatory variables.

The distance variable is expected to have a negative relationship with diamond export and is stated as:

DIST = ∑ j = 1 j w j D i j (5)

where,

w i j = Y j Y w (6)

Y j = Country j’s GDP, Y w = World GDP and D i j = Direct distance between country i and country j

Therefore this mathematical equation is thus derived to examine the factors that impact diamond exports:

ln DEx t = β 1 + β 2 ln FDI t + β 3 ln DIST + β 4 ln GDP t + β 5 ln DPx t + β 6 ln REER t + β 7 ln INFRATE t + ε t (7)

where the β’s are the regression parameters to be estimated and the variables as they have been previously defined. ε t is assumed to be the error term.

To illustrate the use of the Augmented Dickey-Fuller tests, consider first an Autoregressive (AR) (1) model:

y t = μ + ρ y t − 1 + ε t (8)

where μ and ρ are parameters and ε t is assumed to be white noise. y is a stationary series if − 1 < ρ < 1 , and y is a nonstationary series if ρ = 1 . If the absolute value of ρ is greater than one, the series becomes explosive. Therefore, we test whether the absolute value of ρ is strictly less than 1 by making the assumption that the null hypothesis is: ρ = 1 and then test it against the alternative hypothesis: ρ < 1.

Subtracting y t − 1 from both sides of Equation (8)

Δ y t = μ + γ y t − 1 + ε t (9)

where γ = ( ρ − 1 ) and the null and alternative hypotheses are H 0 : γ = 0 and an alternate hypothesis of

H 0 : γ < 0 (10)

The ADF approach controls for higher-order correlation by adding lagged difference terms of the dependent variable y to the right-hand side of the regression:

Δ y t = μ + γ y t − 1 + Σ Δ y t − 1 + ε t (11)

This augmented specification is then used to test the following hypothesis:

H 0 : γ = 0 and an alternate hypothesis of H 1 : γ < 0

The averages for diamonds exports and yearly FDI for the period under study are 538,593 carats and $5.1 million respectively. Also values for real exchange were 4.69 GHS^{16} (max) and 0.015 GHS (min); and 59.5% (max) and 8.7% (min) were recorded for the rate of inflation for the period under study.

The ADF tests results show that all the variables to be insignificant at all levels but significant at first difference.

Variable | Mean | Median | S.D | Maximum | Minimum | Kurtosis |
---|---|---|---|---|---|---|

DEx | 538592.7 | 624871.5 | 296855.3 | 1013616 | 62786 | 1.632910 |

FDI | 5092249 | 5157463 | 2616624 | 9588444 | 225789 | 2.011935 |

GDP | 5.396875 | 4.800000 | 2.313283 | 14.00000 | 2.200000 | 7.258947 |

DPx | 31.35125 | 29.32000 | 8.433412 | 56.04000 | 20.20000 | 4.675403 |

REER | 1.154781 | 0.829500 | 1.371977 | 4.690000 | 0.015000 | 3.852173 |

INFRATE | 20.95000 | 16.80000 | 12.05918 | 59.50000 | 8.700000 | 4.738071 |

Observation | 32 | 32 | 32 | 32 | 32 | 32 |

Source: Authors’ Computation.

Correlation | DEx | FDI | GDP | DPx | REER | INFRATE |
---|---|---|---|---|---|---|

DEx | 1.000000 | |||||

FDI | 0.731040 | 1.000000 | ||||

GDP | −0.158474 | 0.072525 | 1.000000 | |||

DPx | −0.512525 | −0.365360 | 0.438818 | 1.000000 | ||

REER | −0.402260 | −0.451726 | 0.205720 | 0.503777 | 1.000000 | |

INFRATE | 0.212771 | 0.262046 | −0.439824 | −0.553409 | −0.576693 | 1.000000 |

Source: Authors’ Computation.

Variable | ADF Stat. | C.V at 1% | C.V at 5% | C.V at 10% | P-Value | T.C |
---|---|---|---|---|---|---|

lnDEx | −0.964937 | −2.641672 | −1.952066 | −1.610400 | 0.2917 | I (1) |

lnFDI | −1.052329 | −2.641672 | −1.952066 | −1.610400 | 0.2576 | I (1) |

lnGDP | −3.591700 | −2.641672 | −1.952066 | −1.610400 | 0.4460 | I (1) |

lnDPx | 0.080500 | −2.641672 | −1.952066 | −1.610400 | 0.7010 | I (1) |

lnREER | −1.865133 | −2.644302 | −1.952473 | −1.610211 | 0.0601 | I (1) |

lnINFRATE | −0.994658 | −2.641672 | −1.952066 | −1.610400 | 0.2799 | I (1) |

D( lnDEx ) | −5.482075 | −2.644302 | −1.952473 | −1.610211 | 0.0000 | I (0) |

D(lnFDI) | −6.567363 | −2.644302 | −1.952473 | −1.610211 | 0.0000 | I (0) |

D( lnGDP) | −6.735390 | −2.644302 | −1.952473 | −1.610211 | 0.0000 | I (0) |

D(lnDPx) | −5.494568 | −2.647120 | −1.952910 | −1.610011 | 0.0000 | I (0) |

D(lnREER) | −2.388794 | −2.644302 | −1.952473 | −1.610211 | 0.0187 | I (0) |

D(InINFRATE) | −6.282966 | −2.644302 | −1.952473 | −1.610211 | 0.0000 | I (0) |

Source: Authors’ Construct.

Having established the univariate properties of the variables, we went ahead to test to see if any long-run relationship existed among the variables. Hence, a Johansen Cointegration test was performed to determine the cointegrating rank of the model and the number of common stochastic trends that exist among the variables. The results of the Johansen Cointegration test are presented in

At most, there are at least three cointegrating vectors that can be established at a 5% significance level among the six variables indicated by both the Trace and Maximum Eigen value statistics. In order to have a full understanding of the extent of the relationship among the explanatory variables and the response variable, the long run model was assessed with the understanding of how the variables were assumed to have an effect on diamond exports based on the Engel-Granger methodology.

The estimated long-run model is given in Equation (12) below:

D ( ln DEx ) = 3.783 + 0.584 ∗ D ( ln FDI ) − 0.339 ∗ D ( ln GDP ) − 0.915 ∗ D ( ln DPx ) − 0.020 ∗ D ( ln REER ) − 0.331 ∗ D ( ln INFRATE ) (12)

Unrestricted Cointegration Rank Test (Trace) | ||||
---|---|---|---|---|

Hypothesized No. of CE(s) | Eigenvalue | Trace Statistics | 0.05 Critical Value | Prob.** |

None* | 0.952943 | 188.0257 | 95.75366 | 0.0000 |

At most 1* | 0.765178 | 99.39041 | 69.81889 | 0.0000 |

At most 2* | 0.659923 | 57.37149 | 47.85613 | 0.0050 |

At most 3 | 0.400737 | 26.09262 | 29.79707 | 0.1260 |

At most 4 | 0.300231 | 11.24304 | 15.49471 | 0.1970 |

At most 5 | 0.030220 | 0.889881 | 3.841466 | 0.3455 |

Unrestricted Cointegration Rank Test (Maximum Eigenvalue) | ||||
---|---|---|---|---|

Hypothesized No. of CE(s) | Eigenvalue | Max-Eigen Statistic | 0.05 Critical Value | Prob.** |

None* | 0.952943 | 88.63527 | 40.07757 | 0.0000 |

At most 1* | 0.765178 | 42.01891 | 33.87687 | 0.0043 |

At most 2* | 0.659923 | 31.27887 | 27.58434 | 0.0160 |

At most 3 | 0.400737 | 14.84958 | 21.13162 | 0.2996 |

At most 4 | 0.300231 | 10.35316 | 14.26460 | 0.1898 |

At most 5 | 0.030220 | 0.889881 | 3.841466 | 0.3455 |

Variable | Coefficient | Std. Error | T-Stat | Prob. |
---|---|---|---|---|

C | 3.783411 | 1.244919 | 3.039083 | 0.0054 |

lnFDI | 0.584338 | 0.121118 | 4.824552 | 0.0000 |

lnGDP | −0.339262 | 0.287037 | −1.181947 | 0.0050 |

lnDPx | −0.915141 | 0.493188 | −1.855563 | 0.1260 |

lnREER | −0.019576 | 0.068244 | −0.286849 | 0.1970 |

lnINFRATE | −0.331001 | 0.235971 | −1.402719 | 0.3455 |

R-squared | 0.641002 | Mean dependent var | 5.638141 | |

Adjusted R-squared | 0.571964 | S.D. dependent var | 0.324283 | |

S.E. of regression | 0.212160 | Akaike info criterion | −0.095589 | |

Sum squared resid | 1.170312 | Schwarz criterion | 0.179237 | |

Log likelihood | 7.529419 | Hannan-Quinn criter. | −0.004492 | |

F-statistic | 9.284751 | Durbin-Watson stat | 1.035544 | |

Prob(F-statistic) | 0.000036 |

Source: Authors’ Construct.

The p-values indicate that in the long run, only FDI was significant at 5% critical level and thus had a slight impact on diamond exports from Ghana. However, GDP, diamond price, real effective exchange rate as well as inflation rate were found to have no direct positive effect on diamond exports. The long-run static equation shows that 64.1 percent of the explanatory variables can be explained by the response variable. Moreover, the F-statistic of 9.28 is found to be significant as the F-test was used to determine whether the regression equation explained a significantly greater amount of diamond exports for the study period.

Having established the existence of a cointegration relationship among the variables, we proceed to estimate the short-run error correction model to assess equilibrium adjustments by using the disequilibrium estimates from the long-run model. The error correction term (ECT) is given as:

ECT t − 1 = Y t − 1 − β 0 − β 1 X t − 1 − β 2 X t − 1 − ⋯ − β n X t − 1 (13)

where the response variable is Y t − 1 , β 0 is constant and β 1 X t − 1 − β 2 X t − 1 − ⋯ − β n X t − 1 are explanatory variables. The ECT cointegrating equation signifying a long-run relationship among the variables is therefore specified as:

ECT t − 1 = 1.000 ln DEx t − 1 − 0.445 ln FDI t − 1 + 1.0285 ln GDP t − 1 + 3.692 ln DPx t − 1 + 0.002 ln REER t − 1 + 0.423 ln INFRATE t − 1 − 9.437 (14)

Having specified the ECT, we estimated the short-run VECM equation as follows:

VECM : Δ y t = β 0 + ∑ i = 1 n β 1 Δ y t − i + ∑ i = 0 n β 2 Δ x t − i + ⋯ + ∑ i = n n β i Δ n t − 1 + φ z t − 1 + μ t (15)

The results of the VECM is stated in Equation (16) as:

D ( ln GCEx t ) = − 0.079 ect t − 1 − 0.139 ∗ D ( ln DEx t − 1 ) + 0.129 ∗ D ( ln FDI t − 1 ) − 0.177 ∗ D ( ln GDP t − 1 ) + 0.626 ∗ D ( ln DPx t − 1 ) + 0.948 ∗ D ( ln REER t − 1 ) − 0.211 ∗ D ( ln INFRATE ) − 0.101 (16)

To interpret the results of the adjustment coefficients in Equation (16), it can be seen that the previous year’s deviation from long-run equilibrium is corrected in the short-run at an adjustment speed of 7.92%. Also, a percentage change in FDI is associated with a 0.12 percentage increase in diamond exports in the short run. Likewise, a percentage change in diamond prices and real effective exchange rate is also associated with a 0.63 and 0.95 percentage increase in diamond exports. However, a percentage change in GDP and inflation rate are associated with a 0.18 and 0.21 percentage decrease in diamond exports respectively. The results of the short-run estimation are consistent with UNCTAD (2017)^{17} which found that the total volume and value of diamonds exports in Botswana and Sierra Leone alone do not lead to an increase in GDP growth and poverty reduction. More so, all the explanatory variables used in the study do not significantly influence diamond exports from Ghana in the short run.

Variable | Coefficient | Std. Error | T-Statistics | Prob. |
---|---|---|---|---|

C | −0.100556 | 0.052524 | −1.914476 | 0.0687 |

ECT Eqn | −0.079244 | 0.093238 | −0.849909 | 0.4045 |

D(lnDEx(−1)) | −0.138921 | 0.186427 | −0.745179 | 0.4641 |

D(lnFDI(−1)) | 0.128869 | 0.125545 | 1.026476 | 0.3158 |

D(lnGDP(−1)) | −0.176790 | 0.196076 | −0.901641 | 0.3770 |

D(lnDPx(−1)) | 0.626487 | 0.428820 | 1.460957 | 0.1582 |

D(lnREER(−1)) | 0.947547 | 0.528620 | −1.272478 | 0.2165 |

D(lnINFRATE(−1)) | −0.211256 | 0.166020 | −1.402719 | 0.3455 |

R-squared | 0.330308 | Mean dependent var | −0.020535 | |

Adjusted R-squared | 0.117224 | S.D. dependent var | 0.155633 | |

S.E. of regression | 0.146226 | Akaike info criterion | −0.784143 | |

Sum squared resid | 0.470407 | Schwarz criterion | −0.410490 | |

Log likelihood | 19.76214 | Hannan-Quinn criter. | −0.664608 | |

F-statistic | 1.550129 | Durbin-Watson stat | 2.198444 | |

Prob(F-statistic) | 0.202661 |

Source: Authors’ Construct.

The results derived from the estimates used in the international trade export model presents some critically important policy implications for the Government of Ghana, the Ministry of Lands and Natural and allied stakeholders in the diamond sector of Ghana. Numerous research has shown the positive impact of FDI on Ghanaian exports both in the short and long run [

This was because prior to winning their respective tenders to take over the Akwatia Mine, none of them had previously been involved in any diamond activities. In the case of Sapper and Associates, in addition to not having any track record in the diamond industry, they failed to mobilize the required initial deposit of $3.2 million, representing 10% of the agreed total investments. In the case of GCDCL, the company initially conducted feasibility studies on the mining field for three years and decided against using their own funds to undertake the mining project but rather scouting for partners outside of Ghana.

^{8}Most recent amendment to the original Mineral and Mining Law in Ghana was enacted in 2015 as the Mineral and Mining (Amendment) Act, (Act 900).

The diamond sector policy frameworks instituted in Ghana, right from the 1986 Minerals and Mining Law, PNDCL 153 to the present regulations^{18} have not been successfully implemented. It is the recommendation of this research that the Government of Ghana through the DIC should set out strict qualification criteria for prospective bidders to avoid the occurrence of the past takeover failures. The DIC must also conduct a thorough background check on bidders and set out stipulations to punish any divestiture company that falters on delivering on its mandate. Secondly, we recommend that the Ministry of Land and Natural Resources through the Minerals Commission lift the ban or introduce workable regulations on artisanal/small-scale diamond mining in the Akwatia area as it is no secret that it is the artisanal miners that have keep the sector afloat for most of the last decade. Thirdly, we recommend that maximum participation by Government and allied stakeholders should be made a priority if the next divestiture will be a success. Previous governments, dating to the start of the 4^{th} Republic from 1992 till date have not paid much attention to any other mineral besides gold and this has hurt the diamond sector. Lastly, we recommend that government as a matter of necessity, to lessen the financial burden on the next takeover company, take a stake in the equity of the Akwatia Diamond Mine as the debts accrued over the years have served as a deterrent for investments.

The main limitation of this study is in its inability to access the official government documents and relevant data pertaining to the divestiture of the Akwatia Diamond Mine won by Sapper and Associates and GCDCL. This resulted in the inability to properly determine if there were any drawbacks in the divestiture documentation and procedure, thereby affecting the scope of the study.

This study encourages further studies into the divestiture documentation and procedure of the government of Ghana in the disposal of State-Owned Enterprises (SOEs).

The authors declare no conflicts of interest regarding the publication of this paper.

Tian, Z., Kwaw-Nimeson, E. and Onyinah, G.O. (2019) Sparkling Reality: Can Foreign Direct Investments Alone Revive the Akwatia Diamond Mine and Diamond Exports in Ghana?—A Cointegration Analysis. Open Journal of Business and Management, 7, 1846-1861. https://doi.org/10.4236/ojbm.2019.74127