Household Inefficient Electricity Consumption and Its Determinants in Rural Areas

Abstract

Using stochastic frontier analysis (SFA) technique on the main data obtained by stratified random sampling, we have examined the inefficiency of families’ energy use in rural areas in this work. The households’ average efficiency is 77.25%, which suggests that there is a 22.75% underlying inefficiency. The stochastic demand frontier shows that minimum wattage requirements and government subsidies have a considerable beneficial influence on residential power usage. Unexpectedly, the number of living rooms does not cause an increase in electricity usage in rural areas. Overall, we discovered that rural regions’ inefficient electricity use is more a result of supply-side limitations than of demand-side ones.

Keywords

Electricity consumption households inefficiency rural stochastic frontier analysis

Introduction

In several academic fields throughout the world, energy efficiency or efficient energy usage has taken on a considerable amount of connotation. Energy efficiency often refers to using less energy while carrying out energy-dependent tasks. Even though it is simpler to characterise anything as being more or less energy efficient, energy efficiency is the ability to provide a given level of service with less energy or a higher level of service with the same amount of energy input. In fact, it is referred to as ’secret fuel’ or ’invisible power’ (Islam & Hasanuzzaman, 2020) and even ’first fuel’ (Azhgaliyeva, Liu, & Liddle, 2020). By calling for a doubling of the worldwide rate of increase in energy efficiency by 2030, the United Nations has taken the lead through its sustainable development objectives (United Nations, 2015). Due to the impacts and commitments associated with global climate change, it has taken up a significant portion of the policy agenda for decision-makers, with a focus on energy resource reduction, conservation and saving. It is increasingly a factor in industrial and commercial competitiveness, and most importantly, energy security. As a result, sustainability is a core component of energy efficiency frameworks.

Our research is focused on estimating home electricity consumption efficiency and identifying the root causes of inefficiency in India’s rural areas. Notably, India’s home electrification rate in 2019 was 99.99% (Saubhagya, 2019). Electricity is now the main energy source for lighting, heating, entertainment and other electricity-related services in both rural and urban locations, as well as in semi-urban and rural areas. Instead of employing a top-down strategy that relies on secondary data and survey results from government agencies, we have embraced a bottom-up strategy using primary data from homes. According to (ToI, 2021) and (The Economic Times, 2021), the average amount of time that power was available in India was 22 hours in rural regions and 23 hours, 36 minutes elsewhere. In addition, behind the industrial sector, the residential sector consumes the second-highest amount of power overall (IEA, 2018). We specifically picked rural homes because, compared to their urban counterparts, they experience less hours of electrical supply and there aren’t many research on the topic. However, our analysis does not necessarily imply that other sectors, including manufacturing and services among others, are purposefully overlooked in the belief that they are more energy-efficient.

Due to widespread usage and adoption of contemporary electric equipment, the demand for electricity would rise along with the process of rapid expansion and development at both the industrial and domestic levels. Therefore, it is imperative that both urban and rural residents have access to power services. It may be argued that there are two viable solutions to this issue: increasing power output and improving electricity efficiency across the board. The latter solution is really the simplest, most practical and most cost-effective. Increases in power efficiency at the home level, as opposed to a continual process of capacity development that requires investments and expenses, might be a more comprehensive approach to policy. While lowering power costs is a household’s primary goal, various families with the same number of electrical equipment may have varied electricity bill amounts. In actuality, these discrepancies are a product of residence inefficiencies driven on by higher than necessary power use. And numerous socioeconomic conditions, home qualities, regional variances, psychological understanding and the attitude of the household all have a significant impact on this inefficiency. Therefore, cutting back on usage while maintaining the same level of household-required electrical services is the essence of electricity efficiency. As a result, our goal is to calculate the efficiency level of power customers in rural regions with a particular emphasis on identifying the causes of inefficiency. First part of our article is a brief introduction, second part outlines the methodology of our study, third section is our findings and discussion and last section is conclusion with appropriate observations and suggestions.

Methodology, Data Sources and Study Variables

Study Area and Sampling

This study is carried out in Bodoland Territorial Area Districts (BTAD) which lies in the lower Brahmaputra valley of Assam, India. It consists of four districts namely Baksa, Chirang, Kokrajhar and Udalguri with headquarter located in Kokrajhar. The region is a rural area lying along the foothills of the Himalayan range sharing international boundary with Bhutan. The map of the study area is given in Figure 1.

Figure 1.

Map of BTAD.

We mostly depend on the primary data collected from the study area. However, secondary sources have been obtained from census reports, statistical handbooks published by the government of Assam, CMIE reports, NSS Reports, reports by various government agencies like CEA, DERC, IEA, government budgets and economic survey reports. Secondary information of households from each district is obtained from the Primary Census Abstracts 2011 (DCHB, 2011), District Census Handbook for Baksa, Chirang, Kokrajhar and Udalguri districts respectively. We find a negligible area of the rural region which is classified as town area (0.003%) (Appendix 1). Using stratified random sampling, samples were drawn from the study area having electricity connectivity. Information of the household has been collected using schedules. Data on electricity use habits and various socio-economic variables has been collected in order to draw a valid database. For sampling, 2 villages from the most backward regions of the districts were identified¹ purposively. Further 15% of the households from each village were selected randomly giving 322 sample households.

Statistical Tools for Analysis

It is challenging to draw conclusions about energy policy based on the energy consumption to gross domestic product (GDP) ratio or energy intensity since it does not evaluate the degree of ‘underlying energy efficiency’ (Filippini & Hunt, 2011). Therefore, we employed stochastic frontier analysis (SFA), a parametric method first established by (Aigner, Lovell, & Schmidt, 1977), (Battese & Corra, 1977) and (Meeusen & Broeck, 1977), to estimate power consumption efficiency. The SFA technique begins by estimating a function from the underlying relationship between the observable input–output relationships, and then uses the function as a frontier to assess inefficiencies. In addition to allowing for data compatibility, we also took into account chance, measurement mistakes, misspecification of the model, and random noise in the form of omitted variables.

Estimation of energy efficiency/inefficiency has been done in the past by various researchers using SFA such as (Haider & Mishra, 2021), (Hu & Honma, 2014), (Adom, 2019), (Adoma, Amakyeb, Abrokwac, & Quaidood, 2018), (Lin & Du, 2014), (Boyd & Lee, 2019), (Filippini & Hunt, 2011), (Filippini & Hunt, 2012), (Filippini, Evans, & Hunt, 8 The contribution of energy efficiency towards meeting CO2 targets, 2013), (Filippini, Hunt, & Zorić, 2014) including the most recent study by (Twerefou & Abeney, 2020). In a similar manner, we used the SFA technique to estimate the frontier for power consumption efficiency. The frontier provides the minimal level of energy or electricity service that a household needs based on the aggregate electricity demand function and the SFA approach by (Aigner, Lovell, & Schmidt, 1977). Any deviation from the frontier is deemed inefficient given the household’s socioeconomic status, household dwelling characteristics and other attributes like the type of the house and location of the household. Finally, the methodology used by (Filippini & Hunt, 2011) (Filippini & Hunt, 2012), (Filippini, Evans, & Hunt, 2013), (Filippini, Hunt, & Zorić, 2014) (Filippini & Hunt, 2016), (Broadstock, Li, & Zhang, 2016) and (Twerefou & Abeney, 2020) to estimate energy efficiency is applied to the aggregate household’s input electricity demand function in rural regions is specified as

e c_{i} = f (y_{i}; g s_{i}; s c_{i}; f m_{i}; r t_{i}; m w_{i}; a a_{i}; l r_{i}; e e_{i})

(1)

Where,

$e c_{i}$ is the cost of electricity consumption by a household in a month (Rupees). The price per unit of electricity is the same for all the households; $y_{i}$ refers to the household monthly average income (Rupees); $g s_{i}$ the government subsidy in monthly electricity bill (Rupees); $s c_{i}$ number of school college going persons; $f m_{i}$ number of family members; $r t_{i}$ number of persons above 65 years of age; $m w_{i}$ minimum watts consumed by a household (Watt)^*; $a a_{i}$ housing area (Bigha)^*; $l r_{i}$ number of living rooms; $e e_{i}$ captures the underlying energy efficiency of the rural households.

Description of variables and justification are given in Table 1 and the underlying energy efficiency determinant factors are given in Table 2.

Note: ^* For details of the units taken see footnotes of Table 1.

Now taking the log-log function, Equation 1 is represented as

l n (e c_{i}) = a + b_{1} l n (y_{i}) + b_{2} l n (g s_{i}) + b_{3} l n (s c_{i}) + b_{4} l n (f m_{i}) + b_{5} l n (r t_{i}) + b_{6} l n (m w_{i}) + b_{7} l n (a a_{i}) + b_{8} l n (l r_{i}) + v_{i} + u_{i}

(2)

The efficiency term, ${ee}_{i} = v_{i} + u_{i}$ ; where $v_{i}$ is the symmetric disturbance or noise term assumed to be normally distributed and $u_{i}$ represents the level of underlying energy efficiency. It is interpreted as the indicator of the inefficient use of energy (or energy wasted) by a household. It is a one-sided non-negative $(u_{i} \geq 0)$ random disturbance term assumed to follow truncated normal distribution. Here we follow, single step approach of SFA analysis assuming truncated normal distribution of the inefficiency ( $u_{i}$ ) term as two step approach produces biased results (Details of inefficiency term and assumption of its components are given in Appendix 2). The level of energy efficiency is expressed as:

{ELF}_{i} = \frac{E_{i}^{F}}{E_{i}} = exp (- \hat{u_{i}})

Where, $E_{i}$ is the observed amount of electricity consumption and $E_{i}^{F}$ is the frontier or minimum demand of electricity of the i^th household and ${ELF}_{i}$ is the electricity consumption efficiency. Additionally, we generated the inefficiency scores of each household, that is, $1 - b c$ ; where $b c$ is the efficiency score obtained from SFA approach using (Battese & Coelli, 1995) estimation. And since we know the total score of efficiency is 1, the remaining of $1 - b c$ gives us the inefficiency score of each household. The inefficiency variables given in Table 2 are then regressed on the inefficiency score of individual households to find the impact of identified inefficiency variables on the individual inefficiency scores of each household. The determinants for electricity consumption inefficiency are econometrically modelled as a multiple regression model given below

I n f_{i} = β_{1} + β_{2} c m_{i} + β_{3} c f_{i} + β_{4} o t_{i} + β_{5} c f l_{i} + β_{6} D_{1} + β_{7} D_{2} + β_{8} D_{3} + β_{9} D_{4} + u_{i}

Where

$I n f_{i}$ is inefficiency score of the households obtained from SFA estimation; $c m_{i}$ is the average electricity cut in minutes per day; $c f_{i}$ is average number of times electricity is cut in day; $o t_{i}$ is the total outstanding bills in a household in rupees; $c f l_{i}$ number of CFL bulbs in a household; $D_{1}$ is the education level dummy; D₁ is the years of formal education, 0 = 10 years of education; 1 = Upto 12; 1= Graduate and above; $D_{2}$ is secondary source of lighting dummy, 0 = Solar; 1 = Inverter; $D_{3}$ is type of floor of household, 0 = Pukka; 1 = Kutcha; $D_{4}$ is roof of the household; 0 = Terrace; 1 = Tin

Study Variables and Its Justifications

Description of the variables along with the justification for its inclusion in the stochastic demand frontier model is given in Table 1. The determinants of efficiency variables are given in Table 2.

Table 1.

Description of Variables and Its Inclusion for Justification with Expected Sign.

Dependent Variable	Definition
Electricity consumption (ec)	Amount of bill for previous month’s electricity consumption (in Rs)²
Government subsidy (gs)	Amount/Portion of money the government bears as a subsidy in consumer’s bill in a month (in Rs)
Household income (y)	Amount of income a household earns per month taking all sources of income combined (in Rs)
Family size (fs)	Total family members residing in the household (number)
School/College (sc)	School and college going students in the household (number)
Retired (rt)	Total retired persons in the household(number)
Minimum watt (mw)	Minimum watt required to run an electric appliance present in the household (watts)³
Area (aa)	Housing area (in bigha⁴)
Living rooms (lr)	Living rooms in the household (number)

Table 2.

Underlying Energy Efficiency Variables (z).

Variable	Definition
Cut in min (cm)	Electricity cut off in minutes in a day (minutes)
Cut frequency (cf)	Average number of times electricity is cut in a day
Outstanding bill⁵ (ot)	Total bill accumulated over a period of time due to payment defaults including surcharges on outstanding bill (in Rs)
CFL bulbs	Total number of CFL bulbs in a household
Education level(el) Upto 10 Upto 12 Graduate and above	Highest level of education obtained by any member resident of the household
Lighting source (ss) Solar Inverter	Which is a better reliable and efficient secondary source of lighting?
Floor type (ft)  d1:Pukka  d2: Kutcha	Type of floors in a household, i.e. Concrete or Mud
Roof type (rt)  d1:Terrace  d2:Tin	Type of roof of the household, i.e. Tin or Terrace

Results and Discussion

Stochastic Demand Frontier

Table 3 provides the statistical properties of the variables used in our model for estimation. The summary statistics examined here are the average, minimum, maximum and the standard deviation of the log values of the variables. From Table 4, the stochastic demand frontier model to hold the presence of inefficiency in household’s electricity consumption is indicated by the non-zero value of $λ$ , that is, 10.78 (where $λ$ is calculated as $λ = \frac{σ_{u}^{2}}{σ_{v}^{2}}$ , and if it equals to zero the model reduces to ordinary least square OLS), while for the idiosyncratic and inefficiency components the standard deviations of the error components are 2.21 and .20 respectively with a joint variance of 4.92. Thus, our stochastic demand frontier model for inefficiency measurement exists.

Table 3.

Descriptive Statistics of Study Variables.

Variable	Mean	Std. Dev.	Min	Max
Dependent variable (output variable)
Current demand	312.4286	315.4243	13	3235
Demand frontier variables (input variable)
Govt. subsidy	51.77019	39.72895	0	236
Income	23990.37	23810.04	3000	300000
Members	4.465839	1.504018	2	12
School college	1.111801	.9763164	0	5
Retired	.9503106	.8149785	0	4
Minimum watt	1356.674	1541.978	11	9049
Area sq. ft.	10285.71	6035.163	3600	28800
Living rooms	3.658385	1.220824	1	8
Determinants of inefficiency variable
Outstanding	1658.345	5184.87	0	64029
Cfl bulbs	5.357143	2.934343	0	15
Cutmin	103.2578	33.4929	50	170
Cut frequency	3.850932	.6445185	2.5	5

Note: Excluding the dummy variables

The stochastic demand frontier result shown in Table 4 shows that the amount of government subsidy (in Rupees) in power consumption granted to the families has had a considerable positive influence on electricity consumption in the rural area. Our results are consistent with the findings of (Banfi, Filippini, & Hunt, 2005), (Rivers & Jaccard, 2011) (Mirnezami, 2014), (Athukorala, Wilson, Managi, & Karunarathna, 2019), who found that government subsidies for gasoline and electricity had a positive effect on their use. This indicates that government subsidies have been successful in increasing household power usage in rural regions.

The household’s minimal wattage consumption has a big influence on how much more power is used. Our findings are consistent with those of (Kavousian, Rajagopal, & Fischer, 2013) , (Yohanis, Mondol, Wright, & Norton, 2008), (Tiwari, 2000), (Bedir, Hasselaar, & Itard, 2013), (Cramer, et al., 1985) and (McLoughlin, Duffy, & Conlon, 2012), which suggest that households with more electric appliances tend to use more electricity.

Table 4.

Electricity Consumption Demand Frontier.

Variable	Coefficient	Std. Error
Govt. Subsidy	.343^*	.038
Income	.020	.069
Members	.020	.144
School college	–.030	.108
Retired	.174	.121
Minimum watt	.120^*	.046
Area	–.046	.064
Living room	–.233^**	.115
Constant	1.713^*	.211
Log likelihood	–124.279
Wald chi-square	125.56^*
$σ_{u}$	2.210
$σ_{v}$	.204
$σ^{2} = σ_{u}^{2} + σ_{v}^{2}$	4.925
$λ$	10.787
Sample size	322

Notre:^*, ^** and ^*** indicate significance at 1, 5 and 10% level of significance respectively.

It is surprising to learn that the number of living rooms in a home has a negative effect on the amount of power used in the study area. Our results suggest that having additional living rooms in a family generally results in lower electricity use in rural areas. Consequently, our findings are in opposition to those of (Brounen, Kok, & Quigly, 2102), (Kavousian, Rajagopal, & Fischer, 2013), (Weismann, Azevedo, Ferrao, & Fernandez, 2011), (Jones & Lomas, 2015) and (Yohanis, Mondol, Wright, & Norton, 2008). As opposed to prior researchers, we counted the number of living rooms, whereas others counted their size, therefore our results may differ. The majority of the distant settlements in rural areas lack additional living rooms in their homes. Therefore, we had regarded all of the household’s rooms as living rooms.

A household’s power use appears to grow by 2% as its average monthly income rises and as the number of members rises. A household’s monthly power consumption rises by 17% when there are more elderly residents living there. Though the results are not significant our findings are in line with studies found by (Brounen, Kok, & Quigly, 2102), (Jones & Lomas, 2015), (Weismann, Azevedo, Ferrao, & Fernandez, 2011), (Yohanis, Mondol, Wright, & Norton, 2008), (Tiwari, 2000), (Bedir, Hasselaar, & Itard, 2013).

Although not statistically significant, our results show that households with more school-age children consume less power overall – by 3% less each month. This may be attributed to students in rural areas using power responsibly since they are aware of how much electricity is wasted in typical household use.

Determinants of Inefficiency

According to Table 5, the average efficiency of power consumption by families in rural areas is 77%, indicating a 23% underlying inefficiency in that sector. Therefore, there is sufficient data to conclude that improving the inefficiency of rural families is a possibility. Beyond this, rather than focusing on factors that affect electricity demand, we are primarily interested in factors that affect inefficient household electricity consumption in rural areas. Using the inefficiency scores of the households from 1-bc (Battese & Coelli, 1995) and the method of efficiency estimation using SFA described by (Jondrow, Lovell, Materov, & Schmidt, 1982), where bc is the efficiency value of each individual household estimated using this method, efficiency scores are calculated.

Table 5 provides the coefficient values and the diagnostic test for the regression analysis. Notably, as inefficiency is the dependent variable, negative coefficients will indicate a reduction in inefficiency or an improvement in efficiency, or vice versa, a reduction in inefficiency would indicate a rise in efficiency.

Our findings indicate that every factor contributing to inefficiency has a detrimental effect on the inefficiency scores for the rural area, suggesting that every factor has the potential to lessen the inefficiency of home power use. Particularly, a decrease in the typical number of power outages per day and a decrease of hours of electricity availability have a considerable detrimental effect on the home inefficiency ratings. This suggests that fewer frequent power outages, or alternatively, longer periods of continuous power, lower the inefficiency of energy usage by households in rural areas.

Table 5.

Determinants of Inefficiency, Rural.

Variable	Coefficient	Std. Error
Cut in minutes	–.1518551^** (–2.00)	.0757562
Cut frequency	–.3997801^* (–3.04)	.1314955
Outstanding	–.0067504 (–1.04)	.00651
CFL bulb	–.0489375 (–1.22)	.0399851
(Ref; L1) EducationL2	–.0309891 (–1.07)	.0290678
EducationL3	–.037313 (–1.34)	.0278774
(Ref; Floor1) Floor2	–.11592^* (–4.62)	.0250862
(Ref; Roof 1) Roof2	–.0852019 (–1.44)	.0592056
(Ref;Lighting1) Lighting2	–.0766316^** (–2.05)	.0373254
Constant	.9122427^* (5.61)	.1626546
R²	.2310
Adjusted R²	.2075
VIF	1.42
Mean efficiency	.7725

Note:^*, ^** and ^*** indicate significance at 1, 5 and 10% level of significance respectively. t Values are in parenthesis.

When compared to homes with kutcha flooring, among other factors, homes with pukka floors significantly reduce a household’s electrical inefficiency. For homes with pukka flooring, inefficiency might be cut by 11%. Inverters are a secondary source of illumination that may lower the inefficiency of power use by 7% when compared to other secondary sources like solar. Our study complements the findings of (Huebner, David, Hamilton, Chalabi, & Oreszczyn, 2016), (Santin, Itard, & Visscher, 2009), (Steemers & Yun, 2009), (Weismann, Azevedo, Ferrao, & Fernandez, 2011), (Yohanis, Mondol, Wright, & Norton, 2008), (Baldini, Trivella, & Wente, 2018) and (Twerefou & Abeney, 2020) taken as the source of variables that causes inefficiency in electricity consumption.

One of the major issues in the study area is higher outstanding debt, which typically builds up as a result of late electrical bill payments. However, the situation is different in the study area, as indicated and discovered by (Basumatary, 2021), whereby a sizable quantity of unpaid bills was amassed by many rural homes who were listed as having pending electrical bills by the electricity department from the time of their initial connection up until the present. Rural energy users frequently complain about these inconsistencies in electricity billing systems. This really cuts down on actual power use, which in turn decreases home inefficiency. A household will likely be more careful in utilising the electrical service by cutting down on power waste in an effort to further minimise his overdue payment. Inevitably, having additional CFL bulbs might further cut a household’s inefficient power use by over 5%. Last but not least, according to the findings of (Kostakis, 2020) and (Broadstock, Li, & Zhang, 2016), households with higher levels of education could also invariably reduce electricity consumption efficiency in rural areas.

Conclusion and Policy Implication

The minimum watt consumption by a home together with monthly power bill subsidies initially considerably increased electricity use in rural regions. As was anticipated, there is sufficient data to demonstrate that rural energy customers exhibit some degree of inefficiency. When the supply and demand sides of the energy service are separated, inefficiency is mostly caused by supply side restrictions, such as the prolonged length of electricity outages and the frequent outages that occur in rural regions. This inefficiency may be reduced by 39 and 15%, respectively, by reducing the cut frequency and cut duration. By promptly and urgently addressing issues with frequent power outages, providers might minimise inefficiencies as a whole. Particularly rural regions are still vulnerable to bare wires swinging down trees and bamboos, endangering the lives of individuals who live near electrical lines and harming transformers for the supply authority. As a result, the present conventional manner of energy distribution has to be modernised through infrastructure investments in the rural electrification network.

Demand-side efficiency might be decreased by households residing in pukka homes and by implementing a dependable supplemental lighting source. For lowering inefficiency in rural regions, it is true that current government housing program like the Indira Awas Yojana (IAY) and the availability of affordable CFL bulbs, as well as knowledge of the reasons of home inefficiency, might be addressed. On a more general note, as was already said, rural households have been greatly burdened and shocked by unpaid household expenses, which suddenly descended upon them like a barrage.

Nevertheless, their need for energy hasn’t been negatively impacted, showing that households in rural regions are well equipped to use electrical services despite all difficulties. On the basis of the date of their initial connection, a significant portion of the houses were rightfully assessed unpaid debts even though the energy department itself had not yet placed electric meters. After that, as they began to receive payments, a sizable lump sum of the unpaid account was levied on such households. It might be argued that further subsidising some or all of the unpaid invoices via careful examination and installation of smart electric meters would significantly benefit rural families. This might result in a significant shift since rural residents could preserve the money generated by precise billing and utilise it to upgrade their dwellings, switch to more efficient power equipment for lighting, and otherwise reduce inefficiency. By placing more emphasis on the characteristics highlighted, it may be possible to identify an inherent mechanism that would lower the total inefficiency of energy usage in rural regions.

Appendices

Appendix 1.

Classification of Villages and Towns.

Region	Villages(PCA,2011)	Towns	Households	Electrification Status (%)
Assam	26395 (25372)	214	7102607	100
Kokrajhar	1068 (1053)	4	205214	100
Chirang	508 (501)	2	113087	100
Baksa	690 (679)	2	206371	100
Udalguri	800 (791)	3	214656	100
Rural	3065 (3024)	11	739328	100

Appendix 2. Inefficiency Determinants.

Our model allows for the households on one side to decrease their electricity consumption by adopting electricity efficient appliances and on the other side improve their efficiency by more responsible electricity use. The frontier estimated in Equation 2 gives the minimum level of electricity that is necessary for a household to get their required amount of electricity service. If a household is not on the frontier the measured distance from the frontier/baseline above it is the level of energy efficiency (Filippini & Hunt, 2016). Here, we estimate the inefficiency term $u_{i}$ by including a vector of inefficiency determinants with its distribution assumed as truncated normal. The efficiency term $u_{i}$ is modified by introducing a vector of inefficiency determinants $z_{i}$ along with a random term $ε_{i}$ given as

u_{i} = γ^{,} z_{i} + ε_{i}

(3)

Where, $z_{i}$ is a vector of underlying electricity inefficiency variables given in Table 2. Here variables in $z_{i}$ are assumed to be of linear functional form (Broadstock, Li, & Zhang, 2016) given as:

f (z_{i}) = γ_{0} + γ_{1} z_{1} + γ_{2} z_{2} + γ_{3} z_{3} + \dots

(4)

Following (Battese & Coelli, 1995), assuming stochastic term $v_{i}$ as normally distributed $v_{i} ~ i i d N (0, σ_{v}^{2})$ and inefficiency term $u_{i}$ follows truncated normal distribution $u_{i} ~ N^{+} (γ^{,} z_{i}, σ_{u}^{2})$ , Equation 2 is estimated using single stage maximum likelihood estimation method. Now based on the value of $u_{i}$ following (Jondrow, Lovell, Materov, & Schmidt, 1982) the level of energy efficiency of the region is estimated using the conditional mean of the efficiency term $E [u_{i} | v_{i} + u_{i}]$ .

Endnote

Farthest village from the district block office and nearby town (Approx. 10 km)

Current demand = Number of units consumed * Price per unit + electricity duty + current surcharge + fixed charge – government subsidy.

Watts are calculated as minimum watts required for running electric appliance multiplied by the number of electric appliances present in the house. Standard minimum watts consumed by an appliance accessed from https://letsavelectricity.com/wattage-power-consumption-of-household-appliances/.

Unit of measurement for land area, 1 bigha = 14,400 sq ft. (NREDC, 2020).

Outstanding bill = Arrear principle + Arrear surcharge.

Footnotes

Declaration of Conflict of Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Use of Artificial Intelligence

During the preparation of this work, the authors used Bing.AI and Quillbot in order to search for related studies and paraphrasing respectively. After using this tool/service, the author(s) reviewed and edited the content as needed and takes full responsibility for the content of the publication.

ORCID iDs

Sijousa Basumatary

Konita Basumatary

References

Adom

P. K.

(2019). An evaluation of energy efficiency performances in Africa under heterogeneous technologies. Journal of Cleaner Production , 209, 1170–1181.

Adoma

P. K.

, Amakyeb

, Abrokwac

K. K.

, & Quaidood

(2018). Estimate of transient and persistent energy efficiency in Africa: A stochastic frontier approach. Energy Conversion and Management , 166, 556–568.

Aigner

, Lovell

C. K.

, & Schmidt

(1977). Formulation and Estimation of Stochastic Frontier Production Function Models. Journal of Econometrics , 6, 21–37.

Athukorala

, Wilson

, Managi

, & Karunarathna

(2019). Household demand for electricity: The role of market distortions and prices in competition policy. Energy Policy , 134, 110932.

Azhgaliyeva

, Liu

, & Liddle

(2020). An empirical analysis of energy intensity and the role of policy instruments. Energy Policy , 145.

Baldini

, Trivella

, & Wente

J. W.

(2018). The impact of socioeconomic and beahvioural factors for purchasing energy efficient household appliances: A case study for Denmark. Energy Policy , 120, 503–513.

Banfi

, Filippini

, & Hunt

L. C.

(2005). Fuel tourism in border regions: The case of Switzerland. Energy Economics , 27(5), 689–707.

Basumatary

(2021, May 16). EastMojo . Retrieved September 2, 2021, from https://www.eastmojo.com/top-news/2021/05/16/the-myth-and-reality-of-electricity-connection- in-assams-rural-areas/

Battese

G. E.

, & Coelli

T. J.

(1995). A Model for Technical Inefficiency Effects in a Stochastic Frontier Production Function for Panel Data. Empirical Economics , 20, 325–332.

10.

Battese

G. E.

, & Corra

G. S.

(1977). Estimation of a Production Frontier Model with Application to the Pastoral Zone of Eastern Australia. Australian Journal of Agricultural Economics , 21, 169–179.

11.

Bedir

, Hasselaar

, & Itard

(2013). Determinants of electricity consumption in Dutch dwellings. Energy and Buildings , 58, 194–207.

12.

Boyd

G. A.

, & Lee

J. M.

(2019). Measuring plant level energy efficiency and technical change in the U.S. metal-based durable manufacturing sector using stochastic frontier analysis. Energy Economics , 81, 159–174.

13.

Broadstock

D. C.

, Li

, & Zhang

(2016). Efficiency snakes and energy ladders : A (meta-) frontier demand analysis of electricity consumption efficiency in Chinese households. Energy Policy , 91, 383–396.

14.

Brounen

, Kok

, & Quigly

J. M.

(2102). Residential energy use and conservation: Economics and demographics. European Economic Review , 56(5), 931–945.

15.

Cramer

J. C.

, Miller

, Craig

, Hackett

B. M.

, Dietz

T. M.

, Vine

E. L.

. … Kowalczyk

D. J.

(1985). Social and engineering determinants and their equity implications in residential electricity use. Energy , 10(12), 1283–1291.

16.

DCHB. (2011). Primary Census Abstracts . Assam: Directorate of Census Operatives.

17.

Filippini

, & Hunt

L. C.

(2011). Energy Demand and Energy Efficiency in the OECD Countries: A Stochastic Demand Frontier Approach. The Energy Journal , 32(2), 59–80.

18.

Filippini

, & Hunt

L. C.

(2012). US residential energy demand and energy efficiency: A stochastic demand frontier approach. Energy Economics , 34, 1484–1491.

19.

Filippini

, & Hunt

L. C.

(2016). Measuring persistent and transient energy efficiency in US. Energy Efficiency , 9, 663–675.

20.

Evans

, Filippini

, & Hunt

L. C.

(2013). 8 The contribution of energy efficiency towards meeting CO2 targets. In Fouquet

(eds), Handbook on Energy and Climate Change (pp. 175–223). Cheltenham, UK and Northampton, MA, USA: Edward Elgar.

21.

Filippini

, Hunt

L. C.

, & Zorić

(2014). Impact of energy policy instruments on the estimated level of underlying energy efficiency in the EU residential sector. Energy Policy , 69, 73–81.

22.

Haider

, & Mishra

P. P.

(2021). Does innovative capability enhance the energy efficiency of Indian Iron and Steel firms? A Bayesian stochastic frontier analysis. Energy Economics , 95.

23.

J. L.

, & Honma

(2014). A Comparative Study of Energy Efficiency of OECD Countries:An Application of the Stochastic Frontier Analysis. Energy Procedia , 61, 2280–2283.

24.

Huebner

, David

, Hamilton

, Chalabi

, & Oreszczyn

(2016). Understanding electricity consumption: A comparative contributionn of building factors, socio-demographics, appliances, behaviours and attitudes. Applied Energy , 177, 692–702.

25.

IEA. (2018). Electricity Statistics . Retrieved April 19, 2019, from International Energy Agency: https://www.iea.org/statistics/electricity/

26.

Islam

M. M.

, & Hasanuzzaman

(2020). Chapter 1 - Introduction to energy and sustainable development. In Hasanuzzaman

M. D.

, & Rahim

N. A.

, Energy for Sustainable Development (pp. 1–18). Academic Press.

27.

Jondrow

, Lovell

C. A.

, Materov

I. S.

, & Schmidt

(1982). On the estimation of technical inefficiency in the stochastic frontier production function model. Journal of Econometrics , 19, 233–238.

28.

Jones

R. V.

, & Lomas

K. J.

(2015). Determinants of high electrical energy demand in UK homes: Socio-economic and dwelling characteristics. Energy and Buildings , 101(15), 24–34.

29.

Kavousian

, Rajagopal

, & Fischer

(2013). Determinants of residential electricity consumption: Using smart meter data to examine the effect of climate, building characteristics, appliance stock, and occupants’ behavior. ENergy , 55(15), 184–194.

30.

Kostakis

(2020). Socio-demographic determinants of household electricity consumption: evidence from Greece using quantile regression analysis. Current Research in Environmental Sustainability 1, 23–30.

31.

Lin

, & Du

(2014). Measuring energy efficiency under heterogeneous technologies using a latent class stochastic frontier approach: An application to Chinese energy economy. Energy , 76, 884–890.

32.

McLoughlin

, Duffy

, & Conlon

(2012). Characterising domestic electricity consumption patterns by dwelling and occupant socio-economic variables: An Irish case study. Energy and Buildings , 48, 240–248.

33.

Meeusen

, & Broeck

V. D.

(1977). Efficiency estimation from Cobb– Douglas Production Functions with Composed Error. International Economic Review , 18(2), 435–444.

34.

Mirnezami

S. R.

(2014). Electricity inequality in Canada: Should pricing reforms eliminate subsidies to encourage efficient usage? Utilities Policy , 31, 36–43.

35.

NREDC. (2020). National Real Estate Development Council . Retrieved from http://naredco.in/Article-details.asp?id=590&prYear=2020&xmon=Aug

36.

Rivers

, & Jaccard

(2011). Electric utility demand-side management programs in Canada. The Energy Journal , 32(4-5), 93–116.

37.

Santin

O. G.

, Itard

, & Visscher

(2009). The effect of occupancy and building characteristics on energy use for space and water heating in Dutch residential stock. Energy and Buildings , 41(11), 1223–1232.

38.

Saubhagya. (2019). Ministry of Power, Govt. of India . Retrieved from https://saubhagya.gov.in/

39.

Steemers

, & Yun

G. Y.

(2009). Household energy consumption: a study of the role of occupants. Research on Building Stocks , 37(5-6), 625–637.

40.

The Economic Times. (2021, August 5). Retrieved March 7, 2022, from https://economictimes.indiatimes.com/industry/energy/power/rural-average-electricity-supply-at-22-17-hours-a-day-23-36-hours-in-cities-in-june-2021-power-ministry/articleshow/85066305.cms?from=mdr

41.

Tiwari

(2000). Architectural, Demographic, and Economic Causes of Electricity Consumption in Bombay. JOurnal of Policy Modelling , 22(1), 81–98.

42.

ToI. (2021, August 5). Times of India . Retrieved March 7, 2022, from https://timesofindia.indiatimes.com/india/power-supply-in-rural-areas-up-83-at-22-hours-cities-nearly-all-day-r-k-singh/articleshow/85071488.cms

43.

Twerefou

D. K.

, & Abeney

J. O.

(2020). Efficiency of household electricity consumption in Ghana. Energy Policy , 144.

44.

United Nations. (2015). Department of Economic and Social Affairs . Retrieved from https://sdgs.un.org/goals/goal7

45.

Weismann

, Azevedo

I. L.

, Ferrao

, & Fernandez

J. E.

(2011). Residential electricity consumption in Portugal: Findings from top-down and bottom-up models. Energy Policy , 39(5), 2772–2779.

46.

Yohanis

Y. G.

, Mondol

J. D.

, Wright

, & Norton

(2008). Real-life energy use in the UK: How occupancy and dwelling characteristics affect domestic electricity use. Energy and Building , 40(6), 1053–1059.