Geraldine Fobi

Title of Paper: Is Access to Reliable Improved Water a Determinant of Urban Rental Values?                                                                       Abstract This study uses primary data and applies the hedonic price method (HPM)to provide evidence of consumer’s willingness-to-pay (WTP) for improved domestic water. We compute the mean marginal WTP for access to domestic water supply in residences for both localized and non-localized ordinary least squares (OLS) models. The study finds that basic utilities such as toilet facilities and access to electricity in urban residences are major determinants of rental values, followed by improved access to water supply. Households are prepared to spend 4%-10% of their income to improve their quality of life. Note Localised, changed to localized Underlined sentence shortened by rather By access improved changed to by improved access, sounds better reversed  Keywords: Africa, Urban, Water, Hedonic Pricing Method, Willingness-to-Pay. 1.       Introduction The United Nations (2014) have observed that over half of the world’s population now lives in urban areas. Further projections show that by 2050, over 80 percent of the world’s population increase will take place in Africa. Choutmert et al. (2014b) have indicated that population in urban areas of developing and emerging economies are growing rapidly. They argue further, by citing the United Nations Population Division (2008) as having indicated that the urban population in Sub-Saharan Africa will be at least tripled by the year 2050. Given the poverty levels and water supply constraints in developing countries, population growth has made the demand for water outstrip its supply, hence creating a huge water deficit in most developing countries. The World Health Organization (WHO, 2000) has acknowledged how access to water in developing countries is a problem plaguing millions of people and worsening the plight of the poor. Institutions responsible for supplying water in Ghana are limited to supplying water to all, especially households in urban Ghana. This is mainly a result of production and distribution challenges, continued population growth without proper urban planning, and non-revenue water losses (van Rooijen et al. 2008).  Water production in Ghana since 2002 has continued to increase from 201 million cubic meters to 232 million cubic meters in 2009, however, the corresponding demand has outstripped the volumes of water supply. In 2007, 2008 and 2009, water demand increased continuously from 394, 404 and 415 million cubic meters whilst the supply stood at 218, 223 and 232 million cubic meters respectively (MWRWH, 2009). In the case of daily production and demand, Fuest and Haffner (2007) citing Ghana Water Company Limited (GWCL, 2004), report that the daily demand in 2003 was estimated to be 1.023 million cubic meters while supply was only 593,000 cubic meters. Generally, the effective supply of water in Ghana is 280,000 cubic meters per day while the daily demand of 763,300 cubic meters per day is twice the effective supply, hence creating a supply deficit (Water Aid, 2005). In Accra-Ghana, “[water demands are almost six times higher than the actual capacity of the water supply system in 2007” (Adank et al., 2011, p.12).  This shows why there is excess demand for potable water resources in Ghana. This water supply-gap to a developing country like Ghana is an inhibition towards meeting development goals (for an example, the Sustainable Development Goal 6[1]). Research to inform investment opportunities in our quest to bridge this supply-gap in developing countries is critical and this study contributes in that respect. Access to improved water supply is one of the urban services aside access to toilet, access to electricity, access to transport and information technology that needs to be addressed in spite of the gargantuan urban population growth. These utilities are not without their external benefits to improving quality of life. There have been calls in the literature for urban authorities to ensure the expansion of such services to a greater proportion of the population if not all (see Choumert et al., 2014).  In the wake of rapid population growth and urbanization, weak housing policies, and inadequate provision of essential utilities with its associated challenges in city management; this study presents the household’s economic value of access to improved water supply through the housing market in urban Ghana. This is important as it seeks to inform policy makers and stakeholders in urban water supply and housing resource allocations. It further provides estimates relevant for cost- benefit analysis of new infrastructure investments. Also, Malpezzi (1985) and Choumert et al (2014; 2016) have observed that only a few studies have been done in this respect particularly in Africa. This study contributes to the literature on housing markets and improved water valuation. The rest of this study is structured as follows: Section 2 presents the literature of the study. A description of the data and fieldwork issues are presented in section 3. The estimation and discussion of results are undertaken in section 4. Section 5 presents the policy relevance and conclusion of the paper. Note The texts in red have been corrected Accordingly 2.      Empirical Literature This section focuses on water valuation studies with HPM application in developing countries. Relative to developed countries, developing countries, especially in Africa lack data on real estate for HPM studies mainly because of porous housing markets. Indeed, most of the HPM studies in developing countries have relied on survey data, and have found that the domestic water supply plays a significant role in determining rental values. (e.g. North and Griffin, 1993; Vand Den Berg & Nauges, 2012; Vásquez, 2013; Choumert et al., 2014, 2016; Amoah & Moffatt, 2017). In a much broader perspective, North and Griffin (1993) used a 1978 random survey of 1,903 households in the Bicol region in the Philippines to investigate the water source as a housing characteristics in determining housing values. They employed the HPM in their WTP study. One of their key findings was that, all income groups value in-house piped-water source highly relative to other home characteristics.   Similarly, Van Den Berg and Nauges (2012) sought to find out how much households were willing to pay for access to the method employed was the HPM. They found a positive and significant impact of access to piped-water on house values. Consistent with existing studies, the estimated WTP value of piped-water connection averaged 5-7% of monthly household expenditure.  Note The has been added to the sentences above and for changed to of Another study that made use of a survey data, is that of Vásquez, (2013). Using 2006 Household Living Standards Measurement Survey data, collected from Guatemala, the author used the HPM to estimate demand for residential water services. The study found a significantly positive impact on residential water services on house values. The estimated value of municipal water services was found to be at least 15 times as much as the average water bill.  Note The underlined sentence edited from of to on Choumert et al. (2014) provided a simplified approach to determining the value of an attribute (access to piped-water) in the housing market using the HPM. In line with our earlier mentioned studies, and others provided in the literature such as Asabere (1981, 2004), Knight et al. (2004), Gulyani & Talukdar (2008) where a positive relationship has been established between attributes which include access to water, and rental rates; they also found a positive and significant relationship between access to piped-water and rental values. They concluded that extending piped-network with a new property increase rental value.  Note I suggest with be added to the text and plural changed to singular in the underlined text above, One very recent study conducted by Amoah & Moffatt (2017) combined the HPM with other valuation methods such as the contingent valuation method (CVM) and travel cost method (TCM) to estimate demand for piped-water services only. The results from the HPM provided a positive and significant relationship between access to piped-water and rental values. Amoah & Moffatt (2017) differ from the current study in that, while they focused on piped-water services, this current study combines both piped & non-piped water services.  Note Differs changed to differ Irrespective of the positive and significant relationship established by most studies, we admit the possibility of also having contrasting results. For example, a proactive policy was implemented by the government of Togo towards halving by 2015, the proportion of people without access to safe drinking water and sanitation, nonetheless, the government has found it difficult to provide these services. Against this background, Choumert et al. (2016) estimated in Dapaong city, the impact of water [and sanitation] access on property values in the housing market. They applied the HPM to establish the traditional relationship between housing values and its attributes. In their best specification model, they found a positive and insignificant impact of piped-water services on house values. Nonetheless, they concluded that access to piped-water has a substantial impact on households. This result suggests that, contrary to the intuitive expectation there is a positive significant relationship between water supply services and rental values, some contrasting strands of results have also been found in empirical studies using HPM (e.g.  Megbolugbe, 1989; Arimah, 1992). Note Results suggest changes to results suggests and a comma added to that 3. Data and Fieldwork Description The study area is the urban areas of the Greater Accra Region (GAR) of Ghana. Ghana is a sovereign country located within the western part of Africa, along the Gulf of Guinea. It covers an area of 238,537 sq. km (equivalent to 92,100 sq. miles). The population of Ghana is estimated to be over 25million. The economy of Ghana is currently facing economic challenges, although in 2011, her annual gross domestic growth rate was as high as 14.1%. It is endowed with a broad range of natural resources which include but not limited to oil, gold, water resources, diamond, and timber. With GAR inclusive, Ghana is divided into ten administrative regions with ten cities or urban areas. GAR contains 16 districts with Accra as its capital city. It has the second highest number of population accounting for about 15% of the entire population. It is also regarded as having the highest population density in Ghana. Most people in this region depend on several sources of water supply for drinking and for general use. This includes pipes, borehole well, sachet and/or bottled water. According to the 2010 Housing and Population Census as provided by the Ghana Statistical Service (2012), the total population of the GAR with 16 districts is 4,010,054. The population in households is 3,888,512 with male and female distributions as 1,938,225 and 2,071,829 respectively. The total number of households is 1,036,426 with an urban household population of 766,955 and a rural household population of 269,471. Since this section of the study focuses on the urban household population, the sample population used by the study is 766,955 households. The sample population included both renters and owners [they were asked how much they will rent their accommodation for], however, only those respondents who could provide rental values were used. In this study, the sampling frame is housing units within each district. The criteria for district selection was that, it should be one of the 16 districts in the GAR of Ghana. The unit of analysis was household level respondents mainly household heads who are 18 years and above, of sound mind, and who fall within the three income brackets (high, medium and low). They should have worked within the last five years and be currently employed or unemployed within the last seven days of the month of the interview. They should be living in the district and should not be visitors, as of the time of the interview. All potential respondents reserved the right to either accept to participate or decline participation in the survey. Being confronted with lack of listing of residents commonly found in developing countries, we followed FAO (2000) recommendation of a case in India as presented by Hadker et al. (1997). By way of application, the field workers interviewed a pre-determined number based on the computed sample size of each district to control for sample size bias. Further, interviewers interviewed urban households distributed around GAR at any time between 8am to 6pm (time hired) within the localities of all 16 districts depending on the availability of sampled respondent. Respondents were clustered according to their respective districts, communities and randomly selected based on our computed sample sizes. Here, they were observed to have an equal probability of being selected in the sample. Also, households (represented by their heads) were observed to have been well represented by district and by community or locality. This study adopts a simplified formula to calculate sample size as developed by Yamane (1967).1 This formula yielded an appropriate sample size of 400 households. However, for the benefits associated with larger sample sizes coupled with availability of resources, we randomly selected a pre-determined number of 1,650 households instead. Twenty-five field workers were used to administer the questionnaire to the 1650 households from March-May, 2014. Hedonic Empirical Model The HPM follows the revealed preference theory, which provides the opportunity to measure, using observed behaviors, the values households attach to various attributes of say a house (see Choumert et al., 2014; Amoah & Moffatt, 2017). The development of the empirical model for this study is sourced from the theoretical model given by Rosen (1974) for hedonic valuation estimations. We adopted a modified version provided by van den Berg and Nauges (2012) to suit our case. This is provided in equation 1. (1) “Where V is the estimated market value of the property (rent per month), S denotes a set of structural, N a set of neighborhood amenities, and W represents the types of water services available (piped-network, other non-piped water sources). The first order derivative of the hedonic price function with respect to one such characteristic yields an indirect estimate of the willingness-to-pay for this particular characteristic”(p.154). This model is adapted and modified to include other variables. Notable in the new variables is neighborhood mean income to capture for socioeconomic characteristics of the neighborhood. All variables in our model are presented in the descriptive statistics in Table 1. 4. Estimation and Discussion of Survey Results Descriptive Statistics The descriptive statistics and the description of the variables used in generating the results as presented in Table 3 are presented in Table 1. Table 1: Descriptive Statistics of Variables Type of Variable Name Description Obs. Mean Std. Dev Min Max Market value of the property (V) Rent per Month in Ghana Cedis (Gh¢) Log_Rent per Month in Ghana Cedis (Gh¢) Water Related Residential Characteristics (W) Continuous Continuous - 138.23 4.39 174.23 0.97 10 2.30 - Access to Improved Reliable Water Supply in Residence (Piped and Non-piped) [+] Dummy- Access to Toilet Facility in Residence [+] Structural/Residential Characteristics (S) Access to Electricity in Residence [+] Number of Garages in Residence [+] Number of Storeroom in Residence [+] Neighbourhood Characteristics (N) Distance to Nearest School (KM)[+] (Primary-University) Distance to Nearest Highway (KM)[+] Mean District Income (Gh¢)[+] Log_Mean District Income (Gh¢) [+] Dummy- Dummy Continuous Continuous Continuous Continuous Continuous Continuous 1648 1648 1648 1648 1648 1648 - 0.10 0.21 0.26 0.65 636.18 6.45 0.20 0.36 0.50 0.47 1.64 89.65 0.14 0 0 0 0.01 0.1 463.62 6.14 1 3 3 9.29 32.14 842.02 6.74 [*] A priori expectation. Exchange Rate used to be 1 Gh¢= 0.319 US$ as at 15/10/2014. We start our analysis by first conducting a non-parametric test (Table 2a) on the null hypothesis that, the rental values of houses with reliable water are not significantly different from those without the reliability of supply. We find that the z-score to be -6.17 (p<0.01), so we reject the null hypothesis and conclude that there is a difference between the ranks of the two cases specified. Variable Obs. Rank-sum Expected Z* Access to Reliable Water Supply in Residence 512 477,205 422,144 -6.17*** No access to Reliable Water Supply in Residence 1,136 881,571 936,632 Combined 1,648 1,358,776 1,358,776 Table 2a: Non-parametric Wilcoxon rank-sum (Mann-Whitney) Test *H0: Rent Value/Month (Access) = Rent Value/Month (No Access) Furthermore, we investigate the differences in rental values amongst different income groups. The Government of Ghana (January, 2016) through the Ministry of Employment and Labor Relations report, current daily minimum wage as Gh¢8 (equivalent of Gh¢240 per month). The monthly household income is categorized into households living below the minimum wage and households living above the minimum wage. We hypothesise that there is no significant difference between rents paid by lower income groups and rents paid by higher income groups. As reported in Table 2b, the z-score is reported as -9.29 (p<0.01), so we reject the null hypothesis and conclude that there is a significant difference between the monthly rent paid by those living below the minimum wage and those living above the minimum wage, as expected. These results satisfy the scope sensitivity test commonly found in valuation studies. Table 2b: Non-parametric Wilcoxon rank-sum (Mann-Whitney) Test *H0: Rent Value/Month (Below Minimum Wage) = Rent Value/Month (Above Minimum Wage) We proceed with our ordinary least square (OLS) estimation for the different functional forms to find out the extent to which rental values for dwellings with access to reliable improved water supply in their residences change relative to those without access. These results are presented in Table 3. Table 3: Estimated Hedonic Regression Results [With (Yes)/Without (No) localization] (1) (2) (3) (4) (5) (6) VARIABLES Model Lin-Log Model Log-Log Model Lin-Log Model Log-Log Model Lin-Log Model Log-Log Access to Water 27.312*** 0.214*** 22.609** 0.187*** 21.879** 0.184*** (9.304) (0.050) (9.349) (0.050) (9.555) (0.051) Access to Toilet 51.808*** 0.492*** 49.160*** 0.477*** 53.097*** 0.491*** (7.936) (0.049) (7.920) (0.049) (7.929) (0.049) Access to Electricity 45.580*** 0.411*** 44.881*** 0.407*** 45.593*** 0.405*** (10.472) (0.095) (10.990) (0.097) (11.400) (0.099) Number of Garage 67.140*** 0.339*** 67.830*** 0.343*** 64.699*** 0.338*** (17.681) (0.083) (17.607) (0.083) (18.002) (0.084) Number of Storeroom 41.443*** 0.205*** 41.470*** 0.205*** 38.834*** 0.196*** (11.592) (0.060) (11.528) (0.060) (11.494) (0.060) Distance to School (Km) -26.777*** -0.154*** -25.370*** -0.146*** -22.053*** -0.130** (8.127) (0.051) (8.306) (0.052) (8.508) (0.053) Distance to Highway (Km) -4.096*** -0.020** -3.989*** -0.019** -4.239*** -0.024*** (1.301) (0.008) (1.273) (0.008) (1.319) (0.008) Mean_District_Income (Log) N/A N/A 126.855*** 0.706*** N/A N/A - - (29.621) (0.158) - - Constant 42.750*** 3.555*** -771.395*** -*** 3.587*** (9.417) (0.092) (189.786) (1.016) (11.679) (0.100) District Dummies No No No No Yes Yes Observations 1,646 1,646 1,646 1,646 1,646 1,646 R-squared Adjusted R-squared Mean VIF AIC BIC-,-,-,328.35 4,-,-,-,310.90 4,-,-,304.73 4,391.22 Dependent Variable: Rent per month in Ghana Cedis (1 Gh¢= 0.319 US$ as at 15/10/2014). aNon-responses reduced the number of observations from 1648 to 1646 in cases where this is reported. Robust standard errors in parentheses, *** p<0.01, ** p<0.05, * p<0.1 In Table 3, six models are estimated in all. We start by estimating two models namely Model 1 (Lin-log) and Model 2 (log-log) without controlling for either district specific effects or mean district income. Second, we included mean-district-income (representing socioeconomic characteristics of districts) and estimate same models, that is the Model 3 (Lin-log) and Model 4 (log-log) because of the heterogeneous nature of the property market and districts in the region. To prevent possible collinearity between mean-district-income and district dummies, we dropped the mean-district-income and included district dummies. All our models are estimated with robust standard errors. This is because cross-sectional data are prone to the problem of heteroskadasticity hence the need to control for it. We observed differences in the results which provides evidence of the importance of the district dummies as shown in model 6. Stated differently, we find that the district dummies provided an improved results evidenced by the co-efficient of variation (R-squared and Adjusted R-squared), Akaike information criterion (AIC) and Bayesian information criterion (BIC). This suggests that the district dummies control for district specific effect better than just the mean-district-income. Furthermore, it is observed that multicollinearity is not severe in all the models estimated. This is because the study used the Variance Inflation Factor (VIF) to test for the severity of multicollinearity. The mean VIFs provide values of 1.10 (Models 3-6) and 1.11 (Models 1&2) respectively, suggesting that the VIFs in all estimated models are not significantly greater than 1 hence multicollinearity is not severe in our models (See Chatterjee and Hadi, 2006; Choumert et al., 2014). The AIC and BIC results further provide evidence that the best model is model 6. Hence we treat model 6 as our preferred model for our discussion. In addition, given that the coefficient of variation is higher in model 6, we use it as our preferred model while we use the others for robustness checks. This study first dropped variables (such as number of kitchens, number of bathrooms, number of guest rooms, access to TV, access to refrigerator, mean-district- savings, etc.) which were highly correlated and/or insignificant in the model and focused on only those variables that had the right signs and significance following the conventional statistical levels provided below the tables. The results support the apriori expectation of a positive and highly significant relationship between water related residential characteristics (access to water supply and access to toilet facilities in residence) and rental values. From our preferred model 6, we find that households with access to domestic water supply and toilet facilities in residence are willing to pay approximately 18% and 49% more in rental values respectively, relative to those without access to these water related utilities. This implies that averagely households in Ghana are willing to pay more in rental values for houses with access to key utilities. This suggests that houses with relatively higher rental values are likely to have access to water supply services and access to toilet facilities. If households are willing to pay, then one can infer that access to these water related utilities are very relevant in Ghana and perhaps other developing countries with similar characteristics. Other recent studies in African context which include but not limited to Amoah (2017), Choumert et al., 2014, 2016; Gulyani and Talukdar (2008); and Knight et al. (2004) had similar findings. Other residential or structural characteristics which are not water related such as electricity, storeroom and garage were also found to be positive and highly significant as expected. We again find households with access to electricity paying about 41% more in rental values compared to those without electricity. Those with garages and storerooms were willing to pay about 34% and about 20% more in rental values, respectively, relative to those without these structural characteristics. Neighborhood characteristics such as distance to school and highway have negative coefficients as expected and are significant. We find that, a one kilometer increase in the distance from the nearest school and highway leads to approximately 13% and 2% decrease in the rental values of the house respectively. Furthermore, in models (3&4), we used the mean district income as a proxy to describe the level of wealth, knowledge, awareness and perception of the neighborhood (see van den Berg and Nauges, 2012). Thus, we expected districts with higher levels of income to have acquired higher levels of knowledge or perhaps are aware of the rationale behind staying in a residence with access to basic utilities (such as toilet and water, electricity etc.), low crime and noise levels etc. to have higher values for their residence. We find that, a one percent change in mean district income changes the price of dwellings by as high as about 71%. Computation of Marginal Willingness-to-Pay (WTP) The study further derived the marginal WTP otherwise called elasticity for access to water supply in their residences. Following van den Berg and Nauges (2012) and Amoah (2017), we adopt the formula without necessarily going through the derivatives just for brevity. This therefore is given as: From the log-log model 2 (without localization) it yields: =23.86%≈24%. In addition, the 95% confidence interval for the estimated parameter ( is 0.12 and 0.37 or 12% and 37%. Moreover, from the log-log model 6 (with localization) it yields: =20.20%≈20%. Also, the 95% confidence interval for the estimated parameter ( is 0.09 and 0.33 or 9% and 33%. In effect, the marginal WTP which represents the elasticity of access to water in GAR shows the degree of responsiveness in rental values as a result of a change in access to water supply. Therefore, access to water elasticity is 0.21 in the case of log-log model (without localization) and 0.18 for log-log model (with localization). In both cases, access to water elasticity lies between 0.18 and 0.21. This shows that potable water is a normal good or better still a necessity. Based on the marginal WTP values, we compute the mean Marginal WTP and its proportion of household income. We find the mean Marginal WTP to be Gh¢ 27.93 per month and this constitutes 4.39% (approximately 4%) of household income. These are shown in Table 4. Table 4: Mean Marginal WTP for Access to Reliable Improved Water Supply in Residence Marginal implicit house value of water per month(GH¢) in Model 2 Marginal implicit house value of water per month(GH¢) in Model 6 (Preferred) Increment as a % of monthly mean district income of Model 6 (Preferred) Increment as a % of Monthly Household Income for Model 6 (Preferred) Mean*2 Mean Mean*3 Mean*4 32.99 [-] 27.93 [-] 4.39% [1.96%-7.17%] 4.39% [1.96%-7.17%] [.] Denote 95% confidence intervals. The study finds a positive attitude towards having a potable reliable water supply in respondents’ residences which is observed through their WTP. This is seen to significantly influence rental values as expected. The estimated WTP is also found within a reasonable household income range (<=5%) described in the literature as “affordable” (See McPhail, 1993; Whittington et al., 1990 and Goldblatt, 1999). It is important to acknowledge that the marginal implicit house value per month (GH¢) should be interpreted as an upper bound as identified by Bartik (1988), Leggett and Bockstael (2000), Choumert et al. (2014) and Amoah (2017). They attributed this to the fact that the utility dummies used in the model may include unobserved attributes and utilities. Therefore, our estimated WTP should be interpreted as upper bound. Interestingly, as shown in Table 5, we find that although respondents are willing to pay for reliable improved water in residence, they are willing to pay more for toilet and electricity in residence than for improved water. This could be explained by the unavailability of immediate substitutes for toilet and electricity in residence compared to water where sachet water is mainly used in the study area. This implies that, policy makers and property owners could focus on making toilet available in residences, followed by electricity then access to improved water in residences. Thus, regarding the determinants for rental values, access to these three utilities are some of the essential determinants of urban rental values in Accra, Ghana. The provision of each of these utilities constitutes about 4%-10% of household monthly take-home income. We admit that given the size of our adjusted coefficient of determination (16%), there could be other key determinants of rental values in GAR not captured in our model. Table 5: Mean Marginal WTP for Toilet and Electricity in Residence Marginal implicit house value for toilet per month(GH¢) in Model 6 (Preferred) Increment as a % of Monthly Household Income for toilet (Model 6: Preferred) Marginal implicit house value for electricity per month(GH¢) in Model 6 (Preferred) Increment as a % of Monthly Household Income for electricity (Model 6: Preferred) Mean*5 Mean*6 Mean Mean*7 63.40 [-] 9.96% [7.62%-12.53%] 49.93 [-] 7.85% [3.68%-12.89%] [.] Denote confidence intervals estimated at 95%. 5. Policy Relevance and Conclusion Policy Relevance In accordance with Sustainable Development Goal 6[1], we provide mean marginal WTP estimates for reliable improved water supply, and other basic utilities such as toilet facilities and electricity in Ghana. This is very important for consideration in any water and sanitation project in Ghana or other developing countries with similar characteristics. It is also relevant to other sectors such as electricity supply which is currently one of the biggest challenges the country faces today. This positive attitude and estimates should serve as a bait to attract private investment into these utility sectors and the housing market. Conclusion The paper argues that lack of access to reliable improved water services misrepresents rental values in urban GAR. To test this hypothesis, the study estimates household’s mean marginal WTP for access to improved water services using the HPM. The study concludes that access to improved water services is an important determinant of rental values in Urban GAR. Nevertheless, respondents value other utilities such as toilet and electricity more than reliable improved water supply. Indeed, households have given enough evidence of willingness to pay higher rental prices for houses with access to essential utilities such as toilet, electricity and water supply. It is therefore recommended that, although the initial cost of having such essential services is very high, homeowners who are able to take the risk now will certainly experience higher demand for their properties. This demand is associated with economic gains to both property owners (higher rental values) and government/District Assembly (property tax values). Lastly, from the perspective of this study, we recommend a further determination of the real cost of such essential services to Ghanaian households so as to evaluate the cost and benefit to inform policy decisions and properly direct resource allocation. ACKNOWLEDGEMENTS: References Adank, M., Darteh B., Moriarty, P., Osei-Tutu, H., Assan D., & van Rooijen, D. (2011). Towards integrated urban Water Management in the Greater Accra Metropolitan Area: Current Status and Strategic Directions for the Future. Resource Centre Network Ghana. Amoah, A. (2017). Demand for domestic water from an innovative borehole system in rural Ghana: stated and revealed preference approaches. Water Policy, 19(1), 46-68. Amoah, A. & Moffatt G.P. (2017). “Estimating Demand for Reliable Piped-Water Services in Urban Ghana: An Application of Competing Valuation Approaches” The School of Economics Working Paper Series, 2017-01, University of East Anglia, UK. Bartik, T. J. (1988). “Measuring the Benefits of Amenity Improvements”, Land Economics, 64(2), pp 172-183. Bateman, I. (1993). Evaluation of the environment: A survey of revealed preference techniques. Center for Social and Economic Research on the Global Environment. Brookshire, D. S., Thayer, M. A., Tschirhart, J., & Schulze, W. D. (1985). A test of the expected utility model: Evidence from earthquake risks. The Journal of Political Economy, 369-389 Chatterjee, S., & Hadi, A. S. (2006). Simple linear regression. Regression Analysis by Example, Fourth Edition, 21-51. Choumert, J., Kéré, N. E., & Laré-Dondarini, A. L. (2016). A Multi-Level Housing Hedonic Analysis of Water and Sanitation Access. Economics Bulletin, 36(2),-. Choumert, J., Stage, J. & Uwera, C. (2014b). Access to Water as a determinant of rental values: A housing hedonic analysis in Rwanda, Journal of Housing Economics, Vol. 26, pp. 48-54. Cummings, R., Schulze, W., Gerking, S., & Brookshire, D. (1986). Measuring the elasticity of substitution of wages for municipal infrastructure: A comparison of the survey and wage hedonic approaches. Journal of Environmental Economics and Management, 13(3), 269-276. FAO [Food and Agriculture Organization of the United Nations, 2000]. Applications of the contingent valuation method in developing countries: A survey, FAO Economic and Social Development Paper 146, Italy. Fuest, V. & Haffner, S. A. (2007). PPP-Policies, Practices and Problems in Ghana’s Urban Water Supply, Water Policy Journal Vol.9, pp 169-192, IWA publishing. Goldblatt, M. (1999). Assessing the effective demand for improved water supplies in informal settlements: a willingness to pay survey in Vlakfontein and Finetown, Johannesburg, Geoforum, 30(1), 27-41. Government of Ghana, Ghana Statistical Service (GSS, 2012). 2010 Population and Housing Census, Summary Report of Final Results, Sakoa Press Limited, Ghana. Government of Ghana (2016).Government announces increment in minimum wage/salaries for  public workers [Accessed on the-@ http://ghana.gov.gh/index.php/media-center/news/2023-government-announces increment-in-minimum-wage-salaries-for-public-workers]. Gulyani, S., & Talukdar, D. (2008). Slum real estate: the low-quality, high-price puzzle in Nairobi’s slum rental market and its implications for theory and practice. World Development, 36(10),-. GWCL [Ghana Water Company Limited, 2004]. Urban Water Supply Coverage – Information provided to the administrator of district assemblies common fund, Accra, Unpublished document. Hadker, N, Sharma, S., David, A. & Muraleedharan (1997). Willingness-To-Pay for Borivli National Park: Evidence from a Contingent Valuation, Ecological Economics, Vol.21 pp 105-122. Knight, J.R., Herrin, W.E., Balihuta, A.M., (2004). Housing Prices and Maturing Real Estate Markets: Evidence from Uganda. Journal of Real Estate Finance and Economics, Vol.28, pp.5-18. Kolbe, J., & Wüstemann, H. (2015). Estimating the Value of Urban Green Space. SFB 649 Discussion Paper-, Technische Universität Berlin, Germany. Leggett, C. G., & Bockstael, N. E. (2000). Evidence of the effects of water quality on residential land prices. Journal of Environmental Economics and Management, 39(2), 121-144. Malpezzi, S., Mayo, S. K., & Gross, D. J. (1985). Housing demand in developing countries. Washington: World Bank. McPhail, A. A. (1993). The “five percent rule” for improved water service: can households afford more?, World Development, 21(6), 963-973. Moffatt, P. G., & Peters, S. A. (2004). Pricing personal services: An empirical study of earnings in the UK prostitution industry. Scottish Journal of Political Economy, 51(5), 675-690. MWRWH [Ministry of Water Resources Works and Housing, 2009]. Water and Sanitation Sector Performance Report, Government of Ghana, Ghana. Nerlove, M. (1995). Hedonic price functions and the measurement of preferences: The case of Swedish wine consumers. European economic review, 39(9),-. North, J. H., & Griffin, C. C. (1993). Water source as a housing characteristic: Hedonic property valuation and willingness to pay for water. Water Resources Research, 29(7),-. Rosen, S. (1974). Hedonic prices and implicit markets: product differentiation in pure competition, Journal of Political Economy. United Nations (2014). The World Population Situation in 2014: A Concise Report, Department for Economic and Social Affairs Population Division, New York. United Nations Population Division (2008). World Urbanization Prospects: the 2007 revision, Department of Economic and Social Affairs, United Nations Population Division, United Nations, New York. Water Aid (May, 2005). National Water Sector Assessment-Ghana. Accessed on the 16/10/2014 @http://www.wateraid.org/~/media/Publications/national-water-sector-assessment-ghana.pdf. Whittington, D., Mu, X.M. and Roche, R. (1990). Calculating the Value of Time Spent Collecting Water - Some Estimates for Ukunda, Kenya. World Development, 18(2): 269-280. WHO [World Health Organization, 2000]. Global water supply and sanitation assessment 2000 report. WHO, United Nations Children’s Fund, Geneva, Switzerland, New York. van den Berg, C., & Nauges, C. (2012). The willingness to pay for access to piped water: a hedonic analysis of house prices in Southwest Sri Lanka. Letters in Spatial and Resource Sciences, 5(3), 151-166. Van-Rooijen, D.J., Spalthoff, D. and Raschid-Sally, L (2008). “Domestic Water Supply in Accra: How Physical and Social Constraints to Planning have greater consequences for the poor. A reviewed paper presented at the 33RD WEDC International Conference, Accra-Ghana on the Theme: Access to Sanitation and Safe Water: Global Partnerships and Local Actions. Vásquez, W. F. (2013). A hedonic valuation of residential water services. Applied Economic Perspectives and Policy, 35(4), 661-678. Yamane, T. (1967). Statistics: An Introductory Analysis, 2nd Ed., New York: Harper and Row. Note The above texts in red have been edited appropriately, text font changed to New Times Roman, text line spacing 1.5, text font 12