Distribution of government residential utility stimulus in Ghana: evidence from the COVID-19 intervention

Figures & data

Figure 1. Individuals responses to the free utility services.

Table 1. Descriptive statistics of the independent variables.

Stats	MS	Sav	Rec	Dec	P	Age	Y	Lift
Mean	0.45	46.23	0.21	0.94	3.55	47.52	3486.12	0.38
Median	0.00	50	0.00	1.00	4.00	44.00	3500.00	0.00
SD	0.50	67.89	0.41	0.23	1.20	11.71	2438.29	0.49
Skewness	0.20	3.39	1.42	−3.85	−0.54	0.49	0.66	0.50
Kurtosis	1.04	17.99	3.01	15.83	2.48	2.73	2.18	1.25
Minimum	0	0	0	0	1	18	355.00	0
Maximum	1	480	1	1	5	74	8000.00	1
N	873	733	840	853	854	848	699	815

Note. Estimates are presented with a maximum of two decimal places.

Source: Authors estimate.

Table 2. Ordered probit regression results.

	Utilities (Water & Electricity)	No	Indifferent	Yes
Models	1	2	3	4
VARIABLES	Co-efficient	Margins (dy/dx)	Margins (dy/dx)	Margins (dy/dx)
Marital Status	−0.5414***	0.1112***	0.0545***	−0.1657***
	(0.148)	(0.031)	(0.016)	(-0.166)
Savings Depleted (during lockdown)	0.0021**	−0.0004**	−0.0002**	0.0006**
	(0.001)	(0.000)	(0.000)	(0.000)
Lockdown Support Recipient==1	0.2270	−0.0468	−0.0229	0.0697
	(0.167)	(0.034)	(0.017)	(0.051)
Support for Lockdown Decision==1	0.7865***	−0.1595***	−0.0782***	0.2377***
	(0.272)	(0.055)	(0.028)	(0.081)
Public Health Knowledge	0.1741***	−0.0357***	−0.0176***	0.0533***
	(0.053)	(0.011)	(0.005)	(0.016)
Age of Respondent (Years)	0.0054	−0.0011	−0.0005	0.0016
	(0.006)	(0.001)	(0.001)	(0.002)
Income of Respondent (Log)	0.1424	−0.0277	−0.0136	0.0413
	(0.122)	(0.028)	(0.012)	(0.037)
Formal Education ==1	2.8314**	−0.5641**	−0.2766**	0.8407**
	(1.297)	(0.267)	(0.131)	(0.392)
Income × Formal Education	−0.3841**	0.0768**	0.0376**	−0.1144**
	(0.164)	(0.034)	(0.017)	(0.050)
Support for Lifting Lockdown Decision==1	0.2234*	−0.0442*	−0.0217*	0.0659*
	(0.129)	(0.026)	(0.013)	(0.039)
Locked Down City Fixed Effects	Yes
/cut1	1.3473
	(0.973)
/cut2	1.8961*
	(0.975)
Wald Chi^2 (13)	57.03***
Pseudo R^2	0.0816
Observations	493

Dep Variable: Preferences for free Water and or Electricity Supplied by the Government.

Robust standard errors in parentheses.

***p < 0.01, **p < 0.05, *p < 0.1.

Table 3. Ordered probit regression results with a median cut-off (randomised data).

	Utilities (Water & Electricity)	No	Indifferent	Yes
Models	1	2	3	4
VARIABLES	Co-efficient	Margins (dy/dx)	Margins (dy/dx)	Margins (dy/dx)
Marital Status	−0.6265***	0.1233***	0.0637***	−0.1870***
	(0.207)	(0.042)	(0.022)	(-0.166)
Savings Depleted (during lockdown)	0.0007	−0.0001	−0.0001	0.0002
	(0.001)	(0.000)	(0.000)	(0.000)
Lockdown Support Recipient==1	0.0429	−0.0084	−0.0044	0.0128
	(0.234)	(0.046)	(0.024)	(0.070)
Support for Lockdown Decision==1	0.9970***	−0.1962***	−0.1014***	0.2976***
	(0.391)	(0.076)	(0.041)	(0.113)
Public Health Knowledge	0.2407***	−0.0474***	−0.0245***	0.0718***
	(0.072)	(0.015)	(0.008)	(0.021)
Age of Respondent (Years)	0.0047	−0.0009	−0.0005	0.0014
	(0.009)	(0.002)	(0.001)	(0.003)
Income of Respondent (Log)	0.1089	−0.0214	−0.0111	0.0325
	(0.172)	(0.034)	(0.018)	(0.051)
Formal Education ==1	3.0005*	−0.5905*	−0.3052*	0.8957*
	(1.815)	(0.360)	(0.185)	(0.536)
Income × Formal Education	−0.3951*	0.0778*	0.0402*	−0.1180*
	(0.229)	(0.046)	(0.023)	(0.068)
Support for Lifting Lockdown Decision==1	0.3192*	−0.0628*	−0.0325*	0.0953*
	(0.184)	(0.037)	(0.018)	(0.054)
Locked Down City Fixed Effects	Yes
/cut1	1.5082
	(1.445)
/cut2	2.1287
	(1.450)
Wald Chi2 (13)	46.90***
Pseudo R2	0.1155
Observations	250

Dep Variable: Preferences for free Water and or Electricity Supplied by the Government.

Robust standard errors in parentheses.

***p < 0.01, **p < 0.05, *p < 0.1.

Table 4. Disaggregated OLS regression results for water and electricity.

	(1)	(2)	(3)	(4)
VARIABLES	Water	Water	Electricity	Electricity
Marital Status	−0.1688***	−0.1655***	−0.1077***	−0.1070***
	(0.057)	(0.057)	(0.040)	(0.040)
Lockdown Support Recipient==1	0.1172**	0.1143**	0.0505	0.0501
	(0.051)	(0.051)	(0.040)	(0.040)
Support for Lockdown Decision==1	0.2561**	0.2538**	0.2755***	0.2751***
	(0.112)	(0.113)	(0.102)	(0.103)
Public Health Knowledge	0.0482**	0.0480**	0.0459***	0.0457***
	(0.020)	(0.020)	(0.015)	(0.015)
Age of Respondent (Years)	0.0034	0.0031	0.0005	0.0004
	(0.003)	(0.003)	(0.002)	(0.002)
Income of Respondent (Log)	0.0718	0.0652	0.0619*	0.0627*
	(0.050)	(0.051)	(0.032)	(0.032)
Support for Lifting Lockdown Decision==1	0.1026**	0.1070**	0.0319	0.0312
	(0.044)	(0.046)	(0.034)	(0.034)
Household Size	0.1862***	0.1769***
	(0.061)	(0.062)
Income × Household Size	−0.0250***	−0.0239***
	(0.008)	(0.009)
Wealth Depleted (during lockdown)	0.0000***	0.0000***
	(0.000)	(0.000)
Savings Depleted (during lockdown)			0.0005***	0.0005***
			(0.000)	(0.000)
Formal Education			1.0699***	1.0701***
			(0.343)	(0.341)
Income × Formal Education			−0.1426***	−0.1427***
			(0.044)	(0.043)
Constant	37.5001***	37.5377***	53.5202***	53.5196***
	(0.446)	(0.458)	(0.278)	(0.276)
Locked Down City Fixed Effects	No	Yes	No	Yes
Observations	365	365	493	493
R-squared (%)	13.00	13.40	10.66	10.67

Dep Variable: Utility Consumption Index.

Robust standard errors in parentheses.

***p < 0.01, **p < 0.05, *p < 0.1.

Table A1. Pairwise correlation matrix.

	MS	Wealth	Support	Lockdown	PE-Education	Age	lncome	HH	Lifting
MS	1
Wealth	0.0395	1
Support	−0.0657	−0.0372	1
Lockdown	−0.0679	0.0236	−0.0541	1
PE-Education	−0.0366	−0.0952	0.0242	0.0267	1
Age	0.5451	0.0264	−0.0098	−0.0734	−0.102	1
Income	0.3867	0.111	−0.1154	0.0222	−0.0382	0.3809	1
HH	0.075	−0.0159	0.0084	−0.0624	−0.0153	0.0349	−0.0887	1
Lifting	−0.0134	0.0379	0.0633	−0.0318	0.1805	0.0167	0.0157	−0.0347	1

Table A2. Pairwise correlation matrix.

	MS	Saving	Support	Lockdown	PH-Education	Age	lncome	F-Education	Lifting
MS	1
Saving	0.0002	1
Support	−0.0650	−0.0283	1
Lockdown	−0.0423	0.027	−0.0448	1
PH-Education	−0.034	0.0051	0.0603	0.0164	1
Age	0.5591	0.0122	−0.0659	−0.0769	−0.0567	1
lncome	0.3642	0.078	−0.1381	0.0287	−0.0242	0.405	1
F-Education	0.2332	−0.0057	−0.0513	−0.0001	0.057	0.282	0.3639	1
Lifting	0.0186	−0.0205	0.0664	−0.0369	0.1187	0.029	−0.0018	−0.0129	1

Data availability statement

Data is available upon reasonable request.