An econometric analysis of domestic fuel consumption in Ghana: Implications for poverty reduction

Figures & data

Table 1. Households access to fuel for cooking in Ghana

Type of cooking fuel use	2003 (%)	2008 (%)	2014 (%)
Electricity	0.37	0.39	0.84
LPG/Natural Gas	7.36	12.05	20.14
Kerosene/Coal/lignite	1.07	0.43	0.18
Wood	63.02	52.05	46.56
Charcoal	26.97	34.82	31.23
Others*	1.20	0.27	1.06
Total households	6,247	11,440	11,362

Note: *straw/shrubs/grass/agricultural crops/animal dung

Source: Author’s computation based on GDHS data 2003, 2008, 2014.

Table 2. Access to lighting in Ghana

Source	Urban			Rural			Ghana
	Accra	Other urban	All	Coastal	Forest	Savannah	All
Electricity	88.1	73.7	78.5	28.9	32.4	16.6	26.9	49.2
Kerosene	8.6	25.2	19.7	70.3	66.8	66.8	71.8	49.3
Gas lamp	0.1	0.4	0.3	0.3	0.4	0.8	0.5	0.4
Candles/flashlight	2.1	0.2	0.8	0.2	0.2	0.5	0.3	0.5
Solar energy	0.1	0.0	0.0	0.0	0.0	0.3	0.1	0.1
Others	1.0	0.6	0.6	0.4	0.3	0.9	0.5	0.6

Source: GLSS data, (2013).

Table 3. Descriptive statistics of socio-economic variables

Variable	(%) Mean	Standard Deviation	Minimum	Maximum
Type of cooking fuel in % (Wood)	46.56	1.178	0	5
Type of fuel for lighting in % (Electricity)	68.61	0.496	0	2
Sex of Household Head in % (Male)	66.56	0.472	0	1
Age of Household Head in Years (in mean)	45.28	15.94	15	95
Number of Household Members (in mean)	3.82	2.497	1	25
Marital Status of Household Head in % (Married = 1, else = 0)	64.69	0.793	0	1
Wealth Status in % (Poorest/Poorer)	42.49	1.395	0	4
Type of Residence in % (Rural)	50.44	0.500	0	1
Educational Level of Household Head in % (Secondary)	46.77	1.010	0	3
Region (in %)
Western	10.94	0.312	0	1
Eastern	11.26	0.316	0	1
Central	10.49	0.306	0	1
Greater Accra	11.20	0.315	0	1
Ashanti	10.6	0.308	0	1
Volta	9.57	0.294	0	1
Brong—Ahafo	11.08	0.314	0	1
Northern	8.59	0.280	0	1
Upper West	7.65	0.266	0	1
Upper East	8.63	0.281	0	1

Source: Authors’ computation.

Figure 1. Energy access for cooking and lighting in Ghana.

Table 4. Marginal effects at representative values (MER) for cooking/multinomial logit

Explanatory Variable	Electricity (1)	LPG/Natural Gas (2)	Charcoal (3)	Wood (4)	Kerosene/Coal/Lignit(5)
Sex of Household Head (Female)	−0.006*** (0.002)	−0.039** (0.019)	0.0444** (0.018)	0.0006 (0.001)	−0.001* (0.001)
Age of Household Head	−0.0001 (0.001)	−0.004*** (0.001)	0.0042*** (0.001)	0.0002*** (0.003)	0.00003 (0.001)
Size of Household	−0.0003 (0.02)	−0.044*** (0.004)	0.0423*** (0.004)	0.002*** (0.003)	0.0001 (0.001)
Residence (Urban)	−0.001 (0.001)	0.223*** (0.033)	−0.2267*** (0.022)	0.006*** (0.001)	−0.001 (0.002)
Educational level of Household Head (Primary)	0.001 (0.001)	0.132*** (0.012)	−0.1299*** (0.012)	−0.003*** (0.002)	0.0001 (0.001)
Marital Status of Household Head (Married)	0.0015* (0.002)	−0.026 (0.019)	0.026 (0.019)	−0.0015* (0.002)	0.001 (0.002)
Wealth Status of Household (Richer)	0.001 (0.002)	0.420*** (0.012)	0.396*** (0.012)	−0.024*** (0.002)	−0.001 (0.001)
Regional Disparity
Western	Base category
Central	−0.001 (0.002)	−0.045 (0.031)	0.05 (0.031)	0.002* (0.011)	−0.001 (0.00)
Greater Accra	0.003* (0.002)	0.036 (0.027)	−0.04 (0.027)	−0.008*** (0.001)	0.0003 (0.001)
Volta	−0.001 (0.001)	−0.082** (0.033)	−0.076** (0.032)	−0.005*** (0.001)	−0.001 (0.00)
Eastern	0.004* (0.002)	−0.118*** (0.033)	0.113*** (0.032)	0.0019 (0.0013)	−0.001 (0.001)
Ashanti	0.002 (0.001)	−0.057** (0.029)	0.058** (0.028)	−0.002 (0.001)	−0.001 (0.001)
Brong-Ahafo	0.003 (0.002)	−0.001 (0.036)	−0.01 (0.035)	0.0084*** (0.002)	−0.0012 (0.001)
Northern	0.006 (0.004)	−0.271*** (0.048)	0.265*** (0.047)	0.001 (0.002)	0.0001 (0.001)
Upper East	0.001 (0.001)	−0.176*** (0.038)	−0.169*** (0.038)	0.01*** (0.001)	0.001 (0.002)
Upper West	0.003 (0.003)	−0.157*** (0.047)	0.159*** (0.047)	0.005*** (0.001)	−0.0003 (0.001)
Number of Observations Prob>Chi^2 Pseudo R^2 Log Likelihood		11,362 0.0000 0.4477 −7165.8996

Note: The standard errors are within brackets; (***) significant at 1 per cent level; (**) significant at 5 per cent level; (*) significant at 10 per cent level. Base category is kerosene.

Table 5. Marginal effects at representative values (MER) for lighting/multinomial logit

Explanatory variable	Electricity	Generator	Others
Sex of Household Head (Female)	0.016*** (0.003)	−0.016*** (0.003)	0.00011 (0.0003)
Size of Household	0.002*** (0.001)	0.0021*** (0.00)	−0.0002*** (0.001)
Age of Household Head	−0.0002* (0.001)	0.0002** (0.001)	−0.000*** (0.001)
Residence (Urban)	0.020*** (0.01)	0.019*** (0.01)	0.001 (0.001)
Educational level of Household Head (Primary)	0.007*** (0.002)	0.008*** (0.002)	−0.001*** (0.001)
Marital Status of Household Head (Married)	0.005 (0.004)	−0.005 (0.004)	0.001** (0.002)
Wealth Status of Household (Richer)	0.002 (0.003)	0.0084*** (0.003)	−0.01*** (0.003)
Regional disparity
Western Base Category
Central	−0.004 (0.005)	0.004 (0.005)	−0.0001 (0.005)
Greater Accra	0.015*** (0.01)	0.013** (0.005)	0.002** (0.002)
Volta	−0.001 (0.005)	−0.001 (0.005)	0.000 (0.005)
Eastern	−0.005 (0.005)	0.004 (0.005)	0.001** (0.001)
Ashanti	−0.006 (0.005)	0.006 (0.005)	0.001* (0.001)
Brong-Ahafo	−0.010* (0.006)	0.009 (0.006)	0.002*** (0.002)
Northern	0.001 (0.01)	0.0003 (0.006)	−0.002*** (0.003)
Upper East	−0.03** (0.01)	0.026** (0.012)	0.0003 (0.002)
Upper West	0.008 (0.01)	−0.006 (0.01)	−0.002*** (0.001)
Number of Observations Prob>Chi^2 Pseudo R^2 Log likelihood		11,362 0.0000 0.3635 −5157.8866

Note: The standard errors are within brackets; ***, ** and * denotes significant levels at 1%, 5% and 10% respectively. Base category is generator.

Table 6. Test for combining outcome categories

Categories tested	Chi^2	df	Prob>Chi^2
Electricity—LPG/Natural gas	46.220	7	0.000
Electricity—kerosene	75.569	7	0.000
Electricity—charcoal	178.934	7	0.000
Electricity—wood	523.481	7	0.000
Electricity—others	403.342	7	0.000
LPG/Natural gas—kerosene	111.070	7	0.000
LPG/Natural gas—charcoal	1455.687	7	0.000
LPG/Natural gas—wood	3372.369	7	0.000
LPG/Natural gas—others	737.483	7	0.000
Kerosene—charcoal	9.092	7	0.246
Kerosene—wood	61.87	7	0.000
Kerosene—others	103.245	7	0.000
Charcoal—wood	2501.032	7	0.000
Charcoal—others	275.720	7	0.000
Wood—others	50.144	7	0.000

Note: H0: All coefficients except intercepts associated with given pair of outcomes are zero (i.e. categories can be collapsed). Source: Authors' estimation.

Table 7. Hausman tests of IIA assumption

Omitted	chi^2	df	P>chi^2	evidence
Electricity	-348.148	15	1.000	for Ho
LPG/Natural gas	-366.007	23	1.000	for Ho
Kerosene	-336.538	22	1.000	for Ho
Charcoal	-337.953	16	1.000	for Ho
Wood	-544.496	16	1.000	for Ho
Others	-321.322	15	1.000	for Ho

Note: H0: Odds (Outcome-J vs Outcome-K) are independent of other alternatives.

IIA = independent of irrelevant alternative. Source: Authors' estimation.

Table 8. Multidimensional energy poverty index (MEPI)

Year	Headcount ratio (H)	Intensity of energy poverty (A)	MEPI (H*A)	Rural/Urban contributions (%)
2003	0.918	0.9998	0.92	53.3	Rural
				46.7	Urban
2008	0.876	0.9277	0.81	55.2	Rural
				44.8	Urban
2014	0.841	0.7780	0.65	57.1	Rural
				42.9	Urban

Source: Author’s computation based on GDHS data 2003, 2008, and 2014