Estimating Household Willingness to Pay for Improved Electricity Supply in Lagos Metropolis

Nigeria’s per capita electricity consumption is one of the lowest in Africa; with the residential sector being the worst hit. The supply from the state owned Power Holding Company of Nigeria (PHCN) is often erratic forcing most households to resort to the use of generators with its attendance noise and air pollution besides other adverse consequences. The study employed non-market valuation techniques (conjoint analysis, contingent method, and multivariate analysis) to assess the Willingness-to-pay for improved electricity supply using two housing estates within Lagos metropolis as case study. The study reveals the attributes of electricity supply services that give the highest utility, and also the Willingness-to-pay bids of the residents of the two estates. These results were combined and tested for sustainability based on the current cost of generating and maintaining electricity by wind turbines which is considered as one of the most cost-effective options available. Simple ‘payback’ analysis shows that the very low Willingness-to-pay bids of the residents of the two estates would barely support completely deregulated private sector participation. By implication, electricity supply to the estates under reference (including similar and estates of inferior socio-economic characteristics in the metropolis) would continue to be provided by the state-owned Power Holding Company of Nigeria (PHCN) at the existing subsidized rates. Original Research Article British Journal of Economics, Management & Trade, 4(5): 672-693, 2014 673


INTRODUCTION
Sub-Saharan Africa is in the midst of a power crisis, marked by insufficient generating capacity, unreliable supplies, high prices, and low rates of access to the electricity grid [1]. The region's capacity for generating power is lower than that of any other region in the world, and growth has generally stagnated. The average price of power in Sub-Saharan Africa is double that of other developing regions, while supply is unreliable [2]. Because new household connections in many countries are not keeping pace with population growth, the electrification rate, already low, is actually declining [2]. The World Bank reckons that 500 million sub-Saharan Africans are without what it calls "modern energy" [3].
Nigeria in particular is facing increasing electricity supply and demand dilemma. In spite of the slow growth in economic activities in recent years, the demand for electricity in Nigeria has continued to increase [4,5]. Nigeria's electricity market, dominated on the supply side by the state-owned Power Holding Company of Nigeria (PHCN) has been incapable of providing minimum acceptable international standards of electricity service reliability, accessibility and availability.
Nigeria's per capita electricity consumption lags behind other African countries. Indeed, Nigeria's per capita energy consumption is 4 times less than the African average and about 19 times less than the world average [6]. Table 1 reveals cross country comparison of energy output. From the table, South Africa with a population of 44.3 million has electricity supply of 30,020 MW which is 12 times what Nigeria with over 160 million has; Egypt with a population of about 77.5 million has 14,250 MW capacities, approximately 6 times higher than Nigeria's [8,7]. Getting electricity in Nigeria requires eight procedures, takes 260 days and costs 873.9% of income per capita, while the country is ranked 178 th out of 185 economies on the ease of getting electricity supply [9].  Okeh (2007) In Nigeria, the residential sector accounts for a greater proportion of the country's electricity consumption with a peak demand of 2,000 GWH per day thereby making it the worst hit sector [10]. With rising population and consequent increase in household sizes across the country, the demand for electricity continues to grow and more power stations would be required to meet both the existing shortage and the increases. The need for more power stations in the country has long been recognized, but most of the attempts at improving supply since the 1970s have not yielded the desired results. High cost of electricity production has been identified as one of the challenges facing the country's power sector [6].
Given the limited and dwindling resources available to the government at various levels, the ongoing reform in the Nigerian power sector is anchored on deregulation. Past experiences have also shown that direct government involvement in procurement has often been marred by ineptitude and inefficiency. The roadmap for the transformation of the power sector is therefore hinged on private sector participation. However, the sustainability of private sector participation is hinged on appropriate pricing and tariff setting. Presently, there is limited knowledge on how much consumers are willing to pay for improved electricity services in the study area.
One popular means of assessing the sustainability of power/electricity supply has been through Willingness-to-pay studies. If people are willing to pay the full costs of a particular service, then it is a clear indication that the service is valued (and therefore will most probably be used and maintained) and that it will be possible to generate the funds required for sustaining and even replicating the project [11,12]. If cost is not recovered through tariffs they must be funded through tax system or declining services will lead to higher cost of selfprovision [13]. History has shown that public subvention cannot be relied upon in times of constrained budgets as this will leave insufficient funds for new investments in operations and maintenance. In the absence of measures of public subsidies, private investment in infrastructure particularly electricity requires that the tariffs reflect the full cost of provision including profit margin as an incentive for the investors [14]. Sustainable energy planning should be done on the basis of cost-effectiveness, continuous availability, unrestricted supply, and satisfactory regard for the environment [15].
Most environmental goods and services cannot be priced in the market. Hence, resort is made to environmental valuation techniques such as the contingent method and the conjoint analysis. The study is therefore a deviation from the usual supply side management approach to infrastructure provision and management. Electricity consumption is a derived demand and Willingness to pay (WTP) is an expression of the demand for a service. It is a strong pre-requisite for cost recovery because it is a measure of user satisfaction of a service and of the desire and capability of users to contribute to its functioning [16]. Studies in Willingness to pay for improved electricity supply have been carried out in a number country, developing and developed [17,18,19,20,21,22,23,24,25,26]. Similar studies have also been carried out across Nigeria [27,28,29]. However, none of these previous Nigerian studies has considered the areas covered by this study, and none has used the approach adopted for this study.
The study is divided into four sections. The introductory section is followed by the history and current state of energy crisis in Nigeria. The third section concentrates on the study methodology and the statistical tools employed for data analysis. The fourth section includes data analysis and discussion of the results as well as the concluding remarks.

Infrastructure and Economic Growth
Cities have played a key role in the evolution of the global economy. Cities are generators of enormous wealth and act as the powerhouse of the national economy. The strength of a nation's economy, the contact point of international economies, the health of our democracy and the vitality of the humanistic endeavors, all are dependent on whether or not the city works [30]. It is also axiomatic that infrastructure is the engine driving the modern cities. The contribution of infrastructure to economic development in general, and to residential properties in particular, is enormous since it provides good neighborhood quality, encourages investment, enhances accessibility, facilitates information flows and helps in achieving highest and best use [31]. For instance, manufacturing, petro-chemical, iron and steel industries; breweries, petroleum refineries, hospitals, schools, barbing saloon, tailoring and welding workshops, to mention but a few, require regular supply of electricity for effective performance of their activities. Besides their direct value as an item in the "consumption basket" of households, infrastructural services are also a means to acquiring other goods and services. The direct and indirect consumption benefits from electricity, for example, include extra hours of study time due to electric lighting, the availability of new forms of entertainment (e.g. cinema, television) and access to labor saving appliances, among others [14].
Unfortunately, the level of deficiencies and the degree of deterioration of infrastructure in Nigerian cities is alarming and worrisome due to inability of government to meet up its social responsibilities to the people [32]. Of the constraints to infrastructural development, funding is pivotal and universal. The depression of the late 80s and 90s which has not abated, has led to substantial reduction in local, state and federal government revenue resulting in abandoned projects, decay and total neglect of infrastructure. At an estimated average cost of US$1.5 million needed to add a Mega Watt of electricity, Nigeria would need a whopping US$150 billion (N18 trillion) to generate additional 100,000MW required for full industrialization of the economy by 2020 based on a growth rate of 13% as projected by the Energy Commission of Nigeria [33]. Previous attempts at improving the funding of infrastructure including electricity supply in Nigeria such as the establishment of Urban Development Bank have made but insignificant change as the bank itself faced funding problem. In the prevailing circumstances of dwindling funds and increasing demand, creative and innovative solutions are required to address the power supply problem in Nigeria. Nigeria's primary energy resources are in excess if her domestic energy requirement such that Nigerians should not experience supply inadequacy on account of quantum of deposit [35]. Nigerian energy resources include gas, crude oil, coal, solar, nuclear, and hydro [36]. However, only four of these resources namely: coal, crude oil, natural gas and hydro, are presently been utilized.

History and Current State of Electricity Supply in Nigeria
Nigeria's power sector currently has less than 8,000 MW of installed capacity, and less than 6 In the 1970s, the industrial sector was the leading consumer of electricity, accounting for over 50 percent of the total, followed by residential consumers. However, since the 1980s, the residential sector has taken over the lead accounting, on the average for over 50 percent of the total (Fig. 1).

Fig. 1. Electricity consumption pattern in Nigeria (1979-2007)
Source: CBN, 2009: Johnson et al. [5] The Nigerian energy industry is probably one of the most inefficient in the world in terms of meeting customers' needs. Currently, the average daily power generation is below 2,700MW, which is far below the peak load forecast of 8,900MW for the currently existing infrastructure [36]. Nigerian energy crisis is in generation, transmission and distribution. These include insufficient funding of power station, obsolete equipment, inadequate generation availability, overloaded transformers, bad feeder pillars, substandard distribution lines, poor/obsolete communication equipments, poor billing system, unwholesome practices/poor customer National Grid. Of the 14 generating plants, 3 are hydro and 11 are thermal (gas/steam). Seven of the fourteen generation stations are over 20 years old. The national grid is made up of 4,889.2km of 330kV lines, 6,319.33km of 132kV lines, 6,098MVA transformer capacity at 330/132kV and 8,090MVA transformer capacity at 132/33kV. The double-digit transmission and distribution losses are extremely large by international standards and are among the highest in the world. The system losses are five to six times what obtain in well-run power systems [33].
Nigeria has made efforts to increase foreign participation in the electric power sector by commissioning independent power projects (IPPs) to generate electricity and sell it to PHCN. In the 1970s, the industrial sector was the leading consumer of electricity, accounting for over 50 percent of the total, followed by residential consumers. However, since the 1980s, the residential sector has taken over the lead accounting, on the average for over 50 percent of the total (Fig. 1).

Fig. 1. Electricity consumption pattern in Nigeria (1979-2007)
Source: CBN, 2009: Johnson et al. [5] The Nigerian energy industry is probably one of the most inefficient in the world in terms of meeting customers' needs. Currently, the average daily power generation is below 2,700MW, which is far below the peak load forecast of 8,900MW for the currently existing infrastructure [36]. Nigerian energy crisis is in generation, transmission and distribution. These include insufficient funding of power station, obsolete equipment, inadequate generation availability, overloaded transformers, bad feeder pillars, substandard distribution lines, poor/obsolete communication equipments, poor billing system, unwholesome practices/poor customer National Grid. Of the 14 generating plants, 3 are hydro and 11 are thermal (gas/steam). Seven of the fourteen generation stations are over 20 years old. The national grid is made up of 4,889.2km of 330kV lines, 6,319.33km of 132kV lines, 6,098MVA transformer capacity at 330/132kV and 8,090MVA transformer capacity at 132/33kV. The double-digit transmission and distribution losses are extremely large by international standards and are among the highest in the world. The system losses are five to six times what obtain in well-run power systems [33].
Nigeria has made efforts to increase foreign participation in the electric power sector by commissioning independent power projects (IPPs) to generate electricity and sell it to PHCN. In the 1970s, the industrial sector was the leading consumer of electricity, accounting for over 50 percent of the total, followed by residential consumers. However, since the 1980s, the residential sector has taken over the lead accounting, on the average for over 50 percent of the total (Fig. 1).

Fig. 1. Electricity consumption pattern in Nigeria (1979-2007)
Source: CBN, 2009: Johnson et al. [5] The Nigerian energy industry is probably one of the most inefficient in the world in terms of meeting customers' needs. Currently, the average daily power generation is below 2,700MW, which is far below the peak load forecast of 8,900MW for the currently existing infrastructure [36]. Nigerian energy crisis is in generation, transmission and distribution. These include insufficient funding of power station, obsolete equipment, inadequate generation availability, overloaded transformers, bad feeder pillars, substandard distribution lines, poor/obsolete communication equipments, poor billing system, unwholesome practices/poor customer relations by staff, low staff morale and lack of regular training [33]. The energy crisis is manifested in blackouts and brownouts. Getting a faulty transformer fixed in most cases is a nightmare, not to talk about replacing one that has been declared 'dead'. More often than not, residents are forced to contribute money to buy transformers, when as a matter of fact, transformers should be provided by the respective PHCN companies. Consumers are confronted with 'crazy' monthly electricity bills irrespective of whether there was power supply or not.
According to a household survey conducted by the Lagos State Central Office of Statistics [34], 85% of sampled households experience power interruption daily; 10% experience it few times a week; 3% experience it few times a month; while 2% do not experience interruption at all. It was also revealed that 48% of the households enjoy between 1-5 hours of electricity supply per day; 30% enjoy supply less than 1 hour per day; 15% are supplied for between 6-10 hours per day and only 2% claimed that to have electricity between 21-24 hours per day. The analysis of the results showed that 90% of the household connected to PHCN supply are dissatisfied with the performance of the authority. Given the epileptic supply, households have to incur extra cost in fueling and maintaining their private generators to augment supply. The use of generators comes with negative externalities such as noise and fumes, beside the running costs and the very hazardous practice of storing petrol meant for the generators in small containers in the homes.

STUDY AREA
This first leg in a series of study of different parts of the metropolis covers two residential housing estates -the LSDPC known as Ojokoro Estate and Millennium Estate, within the same neighborhood in Ifako Ijaiye Local Government area of Lagos. These estates are Lagos State, South-West Nigeria is the smallest of Nigeria's 36 states in terms of land size. Lagos metropolis accounts for 37% of the land mass of Lagos State but hosts about 85% of the population of the state giving an average population density of 20,000 persons per square kilometer. The metropolis' current population is estimated at 17 million which confer on it the status of a mega city and is projected to be the third largest city in the world by the year 2015. In spite of the fact that the seat of the Federal Government has moved from Lagos to Abuja, Lagos metropolis has remained the nerve centre of the nation's commercial, industrial, and property investment activities. Lagos metropolis has the most active property market with the highest average property value and stock of investment [37]. In 2005, Lagos state accounted for 11.7% (N 1, 700.97 billion) of the national GDP second to the oil-rich River state. The state commands 65% of Nigeria's commercial activities, 60% of national industrial investment and foreign trade, 40% of manufacturing value added, 48% of building and construction activities, and 55% of wholesale and retail trade.

Data Collection
The primary focus of this study is to estimate Willingness-to-pay for improved electricity services in Lagos metropolis using two medium-income housing estates developed by the Lagos State Development and Property Corporation (LSDPC) which were thereafter sold to the public some years ago. Primary data was collected through self-administered questionnaires. Systematic random sampling technique was adapted whereby a flat or household in every three blocks is sampled in Millennium Estate; and a flat in every four blocks in Ojokoro Estate, in proportion to the number of blocks in each estate. The study population is all households in the two estates approximated to the number of accommodation units in the estates. The choice of households as the study population is based on the fact that households are the major consumers of urban services; and the state of the infrastructure particularly electricity supply therefore impacts more on households than any other group.
The questionnaire has three sections. The first section deals with household socio-economic characteristics. The second section addresses households' current use of electricity and their Willingness-to-pay bids; while the third section presents profiles of electricity service attributes that the respondents are made to rank. The questionnaires were administered on the heads of households or their spouses as either or both are more likely to be involved with matters relating to family income and expenditure including expenditure on electricity supply. However, in few cases where neither the household head nor the spouse could be reached, questionnaires were administered to any adult member of the household considered sufficiently informed to provide reliable information.

Source: Field Study
According to the statistics in Table 4, electricity supply to the two estates is highly irregular. According to the respondents, power supply In Millennium Estate is characterized by load shedding (59%) and malfunctioning or lack of transformers (33%). For Ojokoro Estate supply is characterized by broken down transformers (43%), load shedding (28%), and faults from the power station (18%). As a result of all these, the two estates suffer frequent power outages. From Table 5, both estates enjoy a maximum of 15 hours of electricity supply a day on the average and a minimum of as low as below 5 hours. Overall, Ojokoro estate enjoys more supply than Millennium estate. In spite of the poor supply, electricity bills appear to be outrageous while the Millennium estate which enjoys lower supply pays more than Ojokoro estate, on the average. A common complaint by electricity consumers in Nigeria is that electricity bills are arbitrary as the meters are rarely read. The use of alternative source of power, that is, private generator is common to both estates as reported on Table 6. On Millennium Estate, 44.23% of the households uses generator for between 1 and 3 days in a week. In Ojokoro Estate, 37.02% uses electricity generators in 1 to 3 days in a week. About 54% of the respondents in Millennium Estate claims to spend between N2, 001 and N5, 000 weekly; while 36% spends between 500 and N2, 000 to run their generators. For Ojokoro Estate, 50% spends between N2, 001 and N5, 000 a week and 12% spends between N500 and N2, 000 weekly.

The Method of Analysis
The study employed the contingent method, conjoint analysis, multivariate analysis and the 'payback period' analysis. Contingent valuation model is based on a hypothetical market situation, which elicits the prices of goods and services which cannot be priced in the market place. It is referred to as contingent because the results are dependent or contingent upon a devised hypothetical market. It is based on the assumption that a properly designed survey can elicit responses comparable to those arising under actual situations. Contingent method entails asking people of either their maximum Willingness-to-pay (WTP) for an increase in environmental quality, or their Willingness-to-accept compensation to forgo such an increase. The binomial logistic regression is one of the various forms of the discrete choice models. The model statistically relates the choice made by an individual to the attributes of that person and the attributes of the alternatives available to the choice maker. The selected attributes of the sampled households and their a priori expectations are as shown in Table 7.
Conjoint analysis is a survey procedure that is used to derive the values of particular attributes of goods and services. Information is collected about individual's choices between different goods that vary in terms of their attributes or service levels. With this information, it is possible to derive values for a particular attribute or service. If price is included as an attribute in the choice scenarios, value can be defined in terms of dollar (or any other currency), which can be used with the valuation approach [38]. There are three different conjoint approaches: conjoint rating, conjoint ranking, and paired comparison.
The payback period decision rule focuses on how long it takes an investor to recover his initial capital outlay; the longer the time, the riskier the investment is considered to be, and vice versa. As a time concept, payback period is particularly useful in assessing certain timerelated risks.

Conjoint analysis
To minimize the confusion that may likely arise if respondents were to evaluate too many profiles, only the four most important attributes were selected namely: monthly electricity consumption, hours of electricity supply per day, preferred type of meter and expected monthly bill. Each attribute was considered on two levels as shown in Table 8. Having unbundled electricity service into 4 attributes on two levels each, 11 different profiles were generated. The total number of possible combination of the four attributes at the chosen levels is 2 X 2 X 2 X 2=16. In this case, the SPSS algorithm generated a fraction of the full concept i.e. eleven profiles. The number of profiles is greater than the minimum number of profiles required which is given by the total number of levels minus total number of attributes plus 1 [39]. Going by this principle and given the selected attributes and their levels in this study, this minimum number is computed as 5. The 11 profiles generated for this study is therefore considered to be more than sufficient. The respondents were made to rank these 11 concept cards which represent different aspects of electricity services as shown in Table 9. The tolerance level of respondents in conjoint studies is between 12-30 concept cards and 6-8 product attributes, depending on the motivation and product awareness of the respondents [40]. This study has used only 11 profiles which is within the tolerance threshold. The number of respondents (483) is adequate. It has been suggested that between thirty and sixty respondents is adequate for investigational work and for developing hypotheses about a market [41]. Given the attributes selected in this study, the overall preference or total utility of a combination (R) of a profile can be expressed by the following mathematical expression: R = U consumption i + U hours j + U metering k + U billing l + constant Where: U consumption i = utility level i of preferred level of consumption U hours j = utility level jof the number of hours of consumption U metering k = utility level k of the type of metering system preferred, U billing l = utility level l of the amount expected as monthly bill Regression analysis was used to determine the marginal utility of each of the selected attributes among the 483 sampled residents of both estates. Weights were attached to each rank to make them continuous. Since there are 11 profiles, the most preferred profile received 11 points; the second most preferred received 10 points, the third 9 points, down to the last, the least preferred, which received 1 point. These weights were taken as the dependent variables while the independent variables were expressed as dummy variables: 1 if available at certain level, and 0 if not available. The result of the regression is shown in Table 10. The model has an adjusted R 2 of 0.472 with an F-value of 44.61 which provides enough grounds to reject the hypothesis (null) that the explanatory variables jointly lack sufficient explanatory power. The regression coefficient of each variable represents the marginal utility derivable from the respective attribute. For example, provision of 1,000 kilowatt hour in a bundle added 1.863 to the total utility while 16-hour and 24-hour supply of electricity adds 1.220 and 1.538, respectively. From this regression, incorporating prepaid metering to service bundle increases total utility by 0.081. Setting utility bill at N2, 000 also increases total utility by 0.720. It is evident from this analysis that the attribute of electricity supply services that gives the highest utility is the availability of 1,000 kilowatt hour to households; while 24-hour and 16-hour supply is second and third, respectively. From the results in Table 11, the first three profiles have 24 hour connection in their mix which goes to reiterate the significance of this attribute.

Contingent valuation
It is important to determine average household expenditure on electricity per month as a basis for comparing Willingness-to-pay bids. Average payment for electricity in a month for Millennium Estate falls within the range N3, 000 to N5, 000 as each flat on this estate receives uniform electricity bills of N4, 600 per month. The situation varies for Ojokoro Estate where the pre-paid metering system is installed. The average electricity bill per month is lower as the respondents claim to pay between N1, 000 and N3, 000. Payment per Kilowatthour of electricity used combined with frequent outages is responsible for the lower figures. Households incur additional expences on private generator for the low amount spent on electricity from the public mains. Table 12 shows the Willingness-to-pay for improved electricity services for residents of the two estates. Out of the 208 households sampled in Millennium Estate, only 38% are willing to pay for improved supply while the remaining 62% believed that they are already paying too much given the poor level of electricity being supplied. They are therefore less willing to pay for any proposed improvement. For Ojokoro Estate, 64% of the sampled households are willing to pay more for improved supply. Though this estate also receives poor supply; the estate is served with prepaid meters which guarantee relatively efficient and transparent billing.

Determinants of Willingness-to-pay for improved electricity supply
It is common in contingent valuation studies to use multivariate techniques in determining how the socio-economic characteristics of the respondents and other variables affect the WTP bids. The purpose of this kind of analysis is two folds. First, if the WTP bids are correlated with the variables suggested by economic theory; this increases the reliability that the bids are indeed the reflections of the respondents' preferences and not just random occurrence. Second, models of the determinants of the respondents' WTP bids can be used to predict how changes in socio-economic and other variables can affect the demand for the good/service.
The estimated Binomial logistic regression model for the two estates is presented on Table 13 with Cox and Snell R 2 of 0.735 and 0.702 for Millennium Estate and Ojokoro Estate, respectively. These results suggest that, at least, 70% of the Willingness-to-pay decision is explained by the selected explanatory variables.

Source: Field Study
The odds ratios are converted into probability by taking the antilog of the coefficients as shown by the exponential beta column i.e. exp (B). The antilog for HHINCOM in Ojokoro Estate is 1.153 which can be interpreted as an increase in household income by one N1, 000 is more than one time likely to increase willingness to pay for improved electricity. In the same vein an increase in household size by 1 is more than two times likely to increase the willingness to pay for improved electricity. An improvement in the level of education the household head, an increase in the age of the household head and the status of the household head as being employed are likely significant determinants of the willingness to pay for improved electricity supply. The number of days private generators is used in a week and the associated cost of running of the generators are likely determinants as well. The number of hours electricity is supplied from the public source (ELECTHOUR) met the a priori expectations. That is, the more the number of hours electricity is supplied from this source, the less the chances that households in the estate are ready to pay for improved electricity supply.
The determinants of willingness to pay for improved electricity supply for the Millennium estate are similar to those for Ojokoro Estate. Household income, household size, level of education, and the age of the household head are likely to influence the decision of household head to pay for improved supply of electricity. For both estates, the more the number of the days that households are made to use their generators due to irregular supply from the public mains the greater the willingness to pay for improved supply. The costs of running generators also influence the willingness to pay.

VIABILITY APPRAISAL -ELECTRICITY GENERATION FROM WIND TURBINE
Though this study is not so much concerned about the appropriate technology to drive the required improvement in electricity supply in the estates under consideration; for the purpose of demonstrating sustainability under a private sector-driven electricity supply however, the payback analysis is carried out. For this purpose, it is assumed that this private sector-driven initiative employs any one of the well known, relatively simple and cost-effective technologythe Wind Turbine. A wind turbine is a device that converts kinetic energy from the wind (wind energy), into mechanical energy by a process known as wind power which may be used to generate electricity (wind turbine) or drive machinery (windmill). The desire to reduce environmental impacts of conventional energy resources has rekindled interest in renewable energy production across the world. Wind turbine produces clean, quiet, renewable and nonpolluting energy [4]. While researches have been undertaken to assess the feasibility of this renewable source of energy in Nigeria [4,41,42] by determining wind speed patterns and potentials, no study has considered its viability to date. Though previous studies have suggested that the northern part of the country offers more potential than the south due to its higher wind speed; the mean wind speed in the south west which is between 3.0 m/s (maximum) and 1.3 m/s (minimum), is sufficient to support small scale electricity generation built on slow running high torque turbines with multi-blade [4,43].
At the current market price, the cost of generating 1 megawatt of electricity required to serve 100 homes through a wind turbine technology is $2,200,000.00. Therefore, approximately 25 megawatt of electricity is required to serve the 2,500 households in Millennium Estate giving a total cost of $55,000,000.00 equivalent to N9,075,000,000.00 at the current exchange rate of 1US$=N165. Combining this with the revenue projection in Table 14, and using the payback period appraisal technique, it will take approximately 692 years for the investment to break even given the Willingness-to-pay bid of the residents estimated at N13,107,744.0 per annum (net From Table 15, a WTP of N75, 000.00 per household per month is required to guarantee a payback period of 6 years which may be considered the acceptable minimum given the uncertain political atmosphere and unstable economic policies, both of which tend to exacerbate the downside risks of investments in the country. A monthly expenditure of N75, 000.00 on electricity for domestic use is certainly beyond the financial capability of most Nigerians, except perhaps, the very affluent, which represent but an insignificant proportion of the population and are to be found in more exclusive residential neighborhood like Victoria Island and Ikoyi.

CONCLUSION AND RECOMMENDATIONS
This study investigates Willingness-to-pay for improved electricity supply in Lagos Metropolis using two medium-income public residential estates as a starting point in a series of similar studies intended to cover different types and composition of residential estates/neighborhoods in the Metropolis.
The results showed that Willingness-to-pay for improved electricity services in Millennium Estate is affected by household income, household size, number of days households use their generators within a week, and the cost of running generator. For Ojokoro Estate, Willingness-to-pay is affected by increase in household income, household size, employment status of household head, number of days generator is used within a week, and the cost of running generator. The Willingness-to-pay bid for Millennium Estate is N4, 936, and N5, 057.47 for Ojokoro Estate. The results of conjoint analysis showed that the attribute of electricity supply service that gives the highest utility to the residents of the two estates is the availability of 1,000-kilowatt hour to each household. Viability appraisal based on simple Payback analysis and using the current cost of installation and operating wind turbines showed that it will take approximately 692 years and 78 years to recoup initial capital for Millennium Estate and Ojokoro Estate, respectively. Cost and revenue projections showed that the residents of the two estates need to be willing to pay substantially more in order to make electricity generation from wind turbines a viable option.
One obvious inference from the study is that sustainable electricity supply to the housing estates under reference through private sector participation is presently not feasible. It implies that electricity supply to the estates and other estates of similar (or lower standards) socio-economic characteristics in the metropolis would continued to be provided by the state-owned Power Holding Company of Nigeria (PHCN) at the existing subsidized rates.
The study throws light on cost recovery in electricity supply. The results therefore provide useful inputs into government policy formulation when choosing and designing alternative source of power supply for residential areas of different socio-economic characteristics in the metropolis particularly in the light of the ongoing private-public partnership arrangements for provision of infrastructure. The results also provide private investors in the energy sector the basis for realistic pricing and tariff design for sustainable power projects, especially when considering wind turbine technology option.