The authors have declared that no competing interests exist.
Economic statistics concerning the quinquennial features of Agriculture employment (A), net Migration (M), Donor aid (D) and Personal remittances (P), available for over forty years from five West African countries have here been related to the GDP (G). The overall results of a multilinear regression (R
Agriculture and employment in agriculture remain a vital source of livelihood for the majority of African countries, as can be observed in Senegal (SEN), Mali (MLI), Gambia (GMB), Guinea Bissau (GNB), and Mauritania (MRT), the five countries on which the study is based. West African agriculture is mainly characterized by subsistence farming, which leads to a dependence on rain-fed agriculture, a low use of irrigation methods, limited public investment in agriculture, gender disparities between males and females, institutional support and a lack of credit facilities to support small and medium sized farmers. All these factors prevent these countries from increasing their agriculture productivity, mitigating and adapting to climate change problems and encouraging agricultural value chains and trade liberalization. According to ILO
The women involved in agriculture in particular face severe challenges. Although they represent 47% of the labor force, they are prominently smallholder farmers, because the patriarchy system has tended to discriminate against them
Migration and remittances are interconnected
The knowledge and statistics about Donor Aid in the ECOWAS region were reviewed and discussed by Engel and Jouanjean
The covariance among migration, employment in agriculture and remittance received from outside, as a component of the GDP, enables the efficiency of foreign aid in West African countries to be disentangled. The present paper explores the roles played by each endogenous variable in the contribution to the GDP of these countries. The study highlights the key features in which the migration of labor from agriculture, remittances received by skilled or unskilled immigrants, and the employment of the young in agricultural industries can contribute toward the economic growth and potential of such countries to achieve sustainable development goals (SDGS).
The aims of the study in particular have been: 1) to build a multilinear regression model (MLR) of the economic statistics available for over forty years from five West African countries by maximizing the Donor aid (D) to GDP (G) (DG) ratio using the ratios of Personal remittances (P) to G (PG), of P to D (PD), and of P to Agriculture employment (A) (PA) as independent variables, and including the net Migration flux (M) in the model; 2) to cluster the five countries, through a principal component analysis (PCA), and to establish a set of partial least square regression models (PLSR) of the five countries, in order to describe a long-term polydromic function that can be used to disentangle Personal remittances, Agriculture employment and migration, to raise the Donor aid-to-GDP ratio.
The data used in this study (
A partial least square regression (PLSR) method is recommended when the number of observations is limited compared to the dependent \ independent variables. This method is used widely in spectroscopy to fit quantitative dimensions
Dataset (
Variables | Comments |
G- GDP Current (US$) | Current GDP |
M- Net Migration | Net migration is the number of immigrants minus the number of emigrants, including citizens and noncitizens, for the five-year period. |
P- Personal Remittances | Personal remittances, received (current US$) |
A- Employment in agriculture (% of total employment) | Employment in agriculture (% of total employment) (modeled ILO estimate) |
D- Net bilateral aid flows | Net bilateral aid flows from DAC donors, Total (current US$) |
Country | Year | G_M$ | M_K | A_% | D_M$ | P_M$ | G/D | P/D_% | Ln(P/A) | PG% |
G | M | A | D | P | GD | PD | PA | PG | ||
GMB | 1982 | 216.05 | 14.88 | 42.13 | 30.31 | 0.19 | 7.13 | 1% | 8.40 | 0.09% |
GMB | 2007 | 1279.70 | -15.44 | 32.92 | 42.44 | 55.66 | 30.15 | 131% | 14.34 | 4.35% |
GMB | 2012 | 1415.01 | -15.44 | 30.54 | 50.85 | 106.35 | 27.83 | 209% | 15.06 | 7.52% |
GMB | 2017 | 1504.95 | -15.44 | 28.48 | 94.89 | 228.18 | 15.86 | 240% | 15.90 | 15.16% |
GNB | 1992 | 226.31 | -30.00 | 73.16 | 64.39 | 1.33 | 3.51 | 2% | 9.81 | 0.59% |
GNB | 1997 | 268.55 | -41.17 | 72.96 | 84.41 | 2.00 | 3.18 | 2% | 10.22 | 0.74% |
GNB | 2002 | 415.84 | -27.93 | 72.51 | 48.1 | 17.63 | 8.65 | 37% | 12.40 | 4.24% |
GNB | 2007 | 695.99 | -17.50 | 71.77 | 88.59 | 43.03 | 7.86 | 49% | 13.30 | 6.18% |
GNB | 2012 | 989.33 | -7.01 | 70.53 | 51.92 | 45.64 | 19.05 | 88% | 13.38 | 4.61% |
GNB | 2017 | 1346.84 | -7.00 | 68.85 | 52.54 | 104.92 | 25.63 | 200% | 14.24 | 7.79% |
MLI | 1977 | 1049.84 | -175.00 | 81.53 | 72.82 | 26.50 | 14.42 | 36% | 12.69 | 2.52% |
MLI | 1982 | 1333.75 | -218.06 | 79.44 | 115.05 | 39.41 | 11.59 | 34% | 13.11 | 2.95% |
MLI | 1987 | 2090.63 | -493.98 | 77.35 | 255.31 | 88.18 | 8.19 | 35% | 13.95 | 4.22% |
MLI | 1992 | 2830.67 | -173.49 | 74.02 | 310.3 | 116.55 | 9.12 | 38% | 14.27 | 4.12% |
MLI | 1997 | 2697.11 | -141.95 | 73.31 | 308.09 | 91.72 | 8.75 | 30% | 14.04 | 3.40% |
MLI | 2002 | 3889.76 | -67.11 | 71.54 | 308.59 | 137.65 | 12.60 | 45% | 14.47 | 3.54% |
MLI | 2007 | 8145.69 | -100.82 | 69.78 | 736.61 | 343.92 | 11.06 | 47% | 15.41 | 4.22% |
MLI | 2012 | 12442.75 | -302.45 | 68.06 | 818.1 | 827.46 | 15.21 | 101% | 16.31 | 6.65% |
MLI | 2017 | 15337.74 | -200.00 | 63.01 | 928.83 | 883.26 | 16.51 | 95% | 16.46 | 5.76% |
MRT | 1977 | 540.64 | -9.70 | 71.42 | 36.07 | 0.31 | 14.99 | 1% | 8.37 | 0.06% |
MRT | 1982 | 750.21 | -16.10 | 69.17 | 75.65 | 2.32 | 9.92 | 3% | 10.42 | 0.31% |
MRT | 1987 | 909.82 | -40.00 | 66.91 | 107.66 | 6.70 | 8.45 | 6% | 11.51 | 0.74% |
MRT | 1992 | 1464.39 | -44.62 | 63.10 | 159.4 | 50.13 | 9.19 | 31% | 13.59 | 3.42% |
MRT | 1997 | 1401.95 | -44.00 | 62.57 | 177.75 | 2.69 | 7.89 | 2% | 10.67 | 0.19% |
SEN | 1982 | 3936.76 | -85.11 | 58.10 | 228.62 | 66.31 | 17.22 | 29% | 13.95 | 1.68% |
SEN | 1987 | 6381.39 | -60.29 | 54.47 | 432.65 | 117.82 | 14.75 | 27% | 14.59 | 1.85% |
SEN | 1992 | 7602.01 | -77.00 | 48.72 | 494.09 | 175.68 | 15.39 | 36% | 15.10 | 2.31% |
SEN | 1997 | 5915.25 | -227.55 | 47.80 | 337.26 | 150.47 | 17.54 | 45% | 14.96 | 2.54% |
SEN | 2002 | 6752.51 | -202.49 | 45.14 | 297.83 | 346.12 | 22.67 | 116% | 15.85 | 5.13% |
SEN | 2007 | 14285.97 | -218.01 | 40.62 | 548.76 | 1193.38 | 26.03 | 217% | 17.20 | 8.35% |
SEN | 2012 | 17825.42 | -214.00 | 36.40 | 803.82 | 1576.23 | 22.18 | 196% | 17.58 | 8.84% |
SEN | 2017 | 21081.67 | -100.00 | 31.54 | 591.98 | 2148.91 | 35.61 | 363% | 18.04 | 10.19% |
Variables | Coef. | SE | Std. Coef. | SE | P | RR |
Intercept | -2.1624 | 4.9389 | 0.665 | |||
PD | 11.4915 | 1.4932 | 1.2893 | 0.1675 | < 0.0001 | 0.699 |
PG | -191.4112 | 39.5330 | -0.8259 | 0.1706 | < 0.0001 | 0.098 |
LN(PA) | 1.2271 | 0.4421 | 0.3881 | 0.1398 | 0.010 | 0.042 |
M | 0.0063 | 0.0066 | 0.0895 | 0.0932 | 0.346 | 0.005 |
0.844 |
The
The
The
The
The
The PCA shown in
The solutions of the complete models are reported in
The coefficient of the three clusters are reported in
Variables | G | M | A | D | GD | GD | P | PG | PD | PA |
G | 1 | -0.375 | -0.463 | 0.877 | 0.41 | -0.411 | 0.94 | 0.488 | 0.644 | 0.868 |
M | -0.375 | 1 | -0.142 | -0.489 | 0.104 | 0.114 | -0.309 | -0.169 | -0.064 | -0.211 |
A | -0.463 | -0.142 | 1 | -0.204 | -0.626 | 0.509 | -0.494 | -0.605 | -0.787 | -0.542 |
D | 0.877 | -0.489 | -0.204 | 1 | 0.088 | -0.265 | 0.723 | 0.357 | 0.339 | 0.583 |
GD | 0.41 | 0.104 | -0.626 | 0.088 | 1 | -0.692 | 0.384 | 0.268 | 0.59 | 0.411 |
DG | -0.411 | 0.114 | 0.509 | -0.265 | -0.692 | 1 | -0.334 | -0.373 | -0.487 | -0.314 |
P | 0.94 | -0.309 | -0.494 | 0.723 | 0.384 | -0.334 | 1 | 0.607 | 0.777 | 0.975 |
PG | 0.488 | -0.169 | -0.605 | 0.357 | 0.268 | -0.373 | 0.607 | 1 | 0.858 | 0.599 |
PD | 0.644 | -0.064 | -0.787 | 0.339 | 0.59 | -0.487 | 0.777 | 0.858 | 1 | 0.812 |
PA | 0.868 | -0.211 | -0.542 | 0.583 | 0.411 | -0.314 | 0.975 | 0.599 | 0.812 | 1 |
Const. | M | A | D | P | PD | PG | LN(PA) | R2 | |
Cl.I | 13.619 | 0.105 | -0.089 | -0.057 | -0.001 | 3.026 | -7.663 | 0.818 | 0.64 |
Cl.II | 5.52 | -0.011 | 0.075 | 0 | 0.002 | 4.753 | -5.383 | -0.232 | 0.58 |
CL.III | 19.711 | 0.019 | -0.142 | -0.017 | 0.002 | 2.779 | 48.822 | 0.693 | 0.99 |
Clusters | M | A | D | P | PD | PG | LN(PA) |
I- GMB, GNB, MRT - E |
0.20 | -0.17 | -0.28 | -0.01 | 0.30 | -0.04 | 0.23 |
II- MLI - |
-0.29 | 0.16 | -0.03 | 0.31 | 0.48 | -0.03 | -0.11 |
III- SEN - |
0.20 | -0.18 | -0.45 | 0.27 | 0.49 | 0.25 | 0.15 |
MLI shows a great benefit from the M and A factors in cluster II (
In short, as far as Migration is concerned, an increase shows a positive response to MLI (Coeff. -0.29:
Importantly, the consequences of a decrease in D appears more favorable to the G-to-D ratio for SEN (-0.45:
The G-to-D ratio is responsible for increasing Personal remises for MLI (0.31) and SEN (0.27)
Clemens et al.
In the present sample of five West African countries, the growth appears more optimistic in the large interval that was considered, that is, at around +2.2%, when considering the G-to-D ratio, and also shows a favorable parabolic trend. However, it should be pointed out that the trend only becomes favorable in the new millennium.
The effects of the migration (as net fluxes) returns on P are delayed over the years, thus the favorable raw Pearson correlations of M with P (-0.309 not significant) cannot account for time-lag effects. According to Rozelle et al
As far as D and M are concerned, the correlation table reflects a general greater effort (D) to deter future migration (r D,M -0.489), where almost all the negative M net fluxes are ligated with a higher income from D. It should be noted that the DM correlation decreases slightly to a value that was recalculated at a parity of G (r D,M.G -0.359). Clemens and Postel
In our analyses, we confirm that, in a long-term framework, migration acts like a magnet for donor aid. However, because of the weak and non-linear influence of migration on the GD ratio, our MLR model puts the partial coefficient of migration in the last position, with a nearly zero coefficient, and accounting for only 0.5% of the variance, thus supporting the independence of the average of G to D from migration.
It should be recalled that the average trend for the GD ratio, in current values, shows increasing values (+0.33 Y-1) with a parabolic acceleration over the years, and the level is somewhat higher for SEN (21.42; +68%
On the other hand, Alemu and Lee
A long-term polydromic function can disentangle migration, agriculture employment and personal remittances to raise the GDP-to-Donor aid ratio in five African countries that have been grouped into three types: