Willingness to Pay for Improved Electricity Service in Nigeria

The data was gathered with the aid of a well-structured questionnaire administered within the cities of Abuja, Ibadan, Port Harcourt and Lagos in Nigeria. The data comprised of three thematic areas. First is the social economic characteristics of the household. Secondly, the nature of the quality electricity supply and how it affects households’ welfare. In the third section, a hypothetical scenario of an improved electricity system that conforms to all the dimensions of quality electricity supply was created. Respondents were asked to state the maximum amount they were willing to pay for such an improved quality of electricity supply system.
The CVM elicitation format that was employed in the study was the discrete choice with a follow-up approach. A first bid was proposed to each respondent. If the respondent agrees to pay that amount, a higher amount was proposed. If he agrees to that, a third amount, higher than the second was further proposed. If he declined to pay the first bid, the follow up bid proposed to the respondent was lower. After going through the follow up process, all respondents were asked to state after careful thoughts what their maximum WTP for the improved electricity service would be. The amounts each respondent states here were compared to the responses from the follow up process to check for consistency.
The Ordered-Probit Model was employed as the main estimation technique for the study. The model estimated using the Ordered Probit regression was:
WTP = β1 HSZ + β2 HY + β3 EDL + β4 REL + β5 CRR + β6 CAP + β7 MO + ε

The model investigates the factors that influence consumers’ willingness to pay (WTP) for the improved electricity service in the study area. In the model, the outcome variable is WTP (coded 1, 2; 1 being N41 – N55 and 2 being Above N55) which is an ordered categorical variable. The variables used as predictors are Household Size (HSZ), Monthly Outages (MO) - which are continuous variables, Household Monthly Income (HY), Highest Educational Level (EDL), Reliability of Current Supply (REL), Cost incurred in damage of appliances (CRR) and Cost of Alternative Power Supply (CAP) - which are categorical variables. However, it should be stated that ‘n-1’ (n being the number of categories) dummies were created for each of the categorical variables in the model. The reference category for Household Monthly Income is Below N51,000, Highest Educational Level is no formal education, Reliability of Current Supply is Excellent, Cost of Damage is Below N2,000.00 and Cost of Alternative Supply is Below N2,000. The models were estimated separately for each of the enumerated cities and full sample for easy comparison.
Microsoft Excel and the STATA statistical package was used in analyzing the collected data.

本研究的数据通过结构化问卷收集，调研区域覆盖尼日利亚阿布贾、伊巴丹、哈科特港及拉各斯四座城市。本次采集的数据涵盖三大主题板块：其一为家庭社会经济特征；其二为当前电力供应的质量现状及其对家庭福利的影响；其三设置了符合全维度优质电力供应标准的理想化升级电力系统场景，要求受访者表明其为该优质电力升级服务愿意支付的最高金额。

本研究采用的条件价值评估法（Contingent Valuation Method, CVM）征询方式为带后续追问的离散选择法。首先向每位受访者提出初始出价，若受访者同意支付该金额，则依次提出更高的第二轮、第三轮出价；若受访者拒绝首轮出价，则随后提出更低的后续出价。完成整套后续追问流程后，再次请受访者经审慎思考后表明其为升级电力服务愿意支付的最高支付意愿（Willingness to Pay, WTP），并将该声明金额与后续追问环节的应答结果进行比对，以检验应答一致性。

本研究以有序Probit模型（Ordered-Probit Model）作为核心估计方法，构建的有序Probit回归模型如下：
WTP = β₁HSZ + β₂HY + β₃EDL + β₄REL + β₅CRR + β₆CAP + β₇MO + ε
本模型旨在探究调研区域内用户为升级电力服务的支付意愿（WTP）的影响因素。模型中的被解释变量为WTP，属于有序分类变量，编码为1和2：其中1代表支付区间为41奈拉至55奈拉，2代表支付金额高于55奈拉。模型所采用的解释变量包括：作为连续变量的家庭规模（Household Size, HSZ）、月度停电次数（Monthly Outages, MO）；作为分类变量的家庭月收入（Household Monthly Income, HY）、最高学历水平（Highest Educational Level, EDL）、当前供电可靠性（Reliability of Current Supply, REL）、家电损坏损失成本（Cost incurred in damage of appliances, CRR）以及替代电源购置成本（Cost of Alternative Power Supply, CAP）。需说明的是，针对每一个分类变量，均生成了n-1个虚拟变量（n为该变量的类别总数）。各分类变量的参照组分别为：家庭月收入低于51000奈拉、无正规学历、当前供电可靠性为优秀、家电损坏损失成本低于2000.00奈拉，以及替代电源购置成本低于2000奈拉。为便于对比分析，本研究分别针对四座调研城市及全样本数据开展了模型估计。

本研究采用Microsoft Excel及STATA统计软件对采集到的数据进行分析。