Non-Technical Electricity Losses

Abstract

Non-technical loss of electricity (comprising theft, fraud, non-payment and billing irregularities) is a significant issue, particularly in developing countries, and represents a large financial burden on utility companies, governments and society as a whole. This paper takes a wholistic and global view of the challenge and provides a broad perspective of the interrelated issues. Media reports and public perception of non-technical losses tend to focus on residential consumers, particularly those with limited financial resources, whereas review of more robust literature indicates that the largest proportion of non-technical losses is often due to industry, state-owned enterprises and relatively well-off residential consumers. Measures to reduce non-technical losses focusing on average residential consumers, such as pre-paid metering, therefore have limited effect on overall losses. Strengthening of legal and regulatory frameworks, particularly with regard to those larger users, and installing high security tamper-resistant metering systems for commercial consumers may have more effect. The reasons for non-technical losses, especially theft, are complex, but the customer–utility relationship is a key determinant. Improvement of this relationship through local participation in development of renewable energy schemes, such as rooftop solar photovoltaics, could bring benefit if challenges such as financing, design of the distribution system, utility company codes and standards and competence in post installation maintenance can be overcome.

1. Introduction

Transmission and distribution (T&D) losses represent energy generated and supplied to the electricity grid but not paid for by consumers. Such losses are features of all electricity systems and result in increased electricity tariffs for customers, increased carbon emissions and in general have a negative impact on the electricity system as a whole. Given the changing and increasing role of electricity as societies strive to decarbonise, for example through the electrification of heating and transport, the efficiency of electricity generation, transmission, distribution and supply is critically important and loss minimisation is crucial to ensure that electricity systems can deliver a clean, secure and affordable energy supply to society.

According to USAID [1], T&D losses are the “canaries in the coal mines” in that they are the primary indicator by which you can quickly judge the financial and operational health of an electricity utility. The International Energy Agency (IEA), using data collected from national energy agencies, tracks and reports on electricity system T&D losses globally [2]. This is a useful tool for comparing losses in particular regions, and the IEA data indicate that globally electric power T&D losses were approximately 8.2% in 2014. Multiple literatures report that T&D losses are more of an issue in developing countries [3,4,5], and according to USAID [1], a well performing utility in a developing country may have a total T&D loss level of c. 10% while T&D losses in developed countries such as the United States average about 5%. The IEA database [2] corroborates this, as shown in Figure 1, which compares T&D losses in the developing regions of India, Latin America and sub-Saharan Africa against T&D losses in the developed regions of the United States, United Kingdom and European Union.

High levels of T&D losses in developing countries present major economic challenges for utilities, and the World Bank reports that utility companies in developing countries often suffer from significant quasi-fiscal deficits, that is the difference between the net revenue of an efficient electricity sector covering operational and capital costs and the net cash collected by the utilities [6]. This deficit comprises four components: (1) T&D losses, (2) bill collection losses, (3) overstaffing and (4) under-pricing. Figure 2 from the World Bank Report, Making Power Affordable for Africa and Viable for its Utilities, Ref. [6] shows that outside of South Africa, T&D losses are the second largest contributor to the quasi-fiscal deficit in Africa, accounting for 30% of hidden costs.

The treatment and categorisation of losses differ from country to country; thus, cross-country comparisons of electrical energy losses are far from straightforward. However, multiple literatures categorise losses as being either technical or non-technical, and total T&D losses is made up of this combination [3,7,8,9]. Technical losses occur as a direct result of the physical characteristics of the electricity network and consist mainly of energy dissipation in electrical system components such as lines, transformers, connections, measurement systems and other equipment that carry energy to and from customers [7]. Non-technical losses are caused by actions external to the power system and consist primarily of electricity theft, non-payment by customers and errors in accounting and record keeping [3]. According to Smith [10], theft can be subdivided into four further categories:

• Fraud—when the electricity consumer deliberately tries to deceive the utility company by, for example, tampering with the electricity meter such that a lower energy consumption (kWh) is shown than is the case.
• Stealing Electricity—this is the practice of bypassing electricity meters by connecting consumers directly to electricity supply points, such as public lighting or a nearby distribution transformer via illegally installed conductors (cables or wires).
• Billing Irregularities—these can be either unintentional or intentional irregularities. An example of an unintentional irregularity would be where the utility company issues incorrect bills to consumers due to a systems failure while an intentional irregularity could be utility company employees knowingly recording meter readings or issuing bills that are less than should be the case.
• Unpaid Bills—where electricity consumers decide not to pay the bill that is owed with the outstanding debt being carried by the utility company.

2. Scale and Impact of Non-Technical Losses

It is difficult to accurately quantify non-technical losses based upon overall T&D losses because, as reported by the European Regulators Group for Electricity and Gas (ERGEG) [11], regulatory definitions of the term “power losses” vary significantly from country to country, and specifically regarding non-technical losses, there is a noteworthy heterogeneity in the definition which hinders the comparison of percentages of losses across countries. For example, while non-technical losses associated with theft and metering error appear to be accounted for by most European utilities, there are differences regarding the treatment of other non-technical losses such as non-metered supply for public lighting or non-metered “internal consumption” by the electricity utility. Furthermore, it is difficult to obtain accurate data associated with non-technical losses, as electricity theft can only be estimated but not measured exactly [10] and some utilities prefer not to disclose loss data, keeping the information private to the company’s sectors and employees [12]. However, it has been estimated that globally non-technical losses could amount to USD 80–100 billion per annum [13,14,15]. Reports of losses in specific countries or regions include:

• During 2011, non-technical losses in Colombia were equivalent to approximately USD 2.0 million per month, which corresponds to 14.7% of the energy generated [12].
• At the end of 2013, the Jamaica Public Service Company Limited (JPS) reportedly lost approximately USD 46 million in revenue (or approximately 18.0% of its total fuel bill) as a result of electricity theft by an estimated 180,000 unmetered consumers [16].
• In 2015, the South African utility ESKOM reported that about USD 350 million worth of electricity was lost to theft [17].
• The Brazilian Electricity Regulatory Agency (ANEEL) estimates that 5% of the energy injected to the distribution grids is lost because of fraud and theft, and from 2017 data, it was estimated that the non-technical losses generated a negative impact of BRL 8.15 billion, approximately EUR 2.3 billion [18].

High levels of non-technical losses present major economic challenges for utilities, and this is especially true for utilities operating in the least developed countries [19]. For example, an email from the CEO of the Liberia Electricity Corporation (P Buckley 2021, personal communication, 20 August) advised that during the period November 2014 to October 2015, approximately 58 GWh of electricity was generated by the Liberia Electricity Corporation (LEC), but only circa 40 GWh was billed to customers This means that approximately 31% of the electricity generated by the utility was “lost” due to a combination of technical and non-technical losses. LEC, with support from aid agencies, developed models and undertook load flow analysis of the electricity system. This enabled LEC to estimate that approximately 7 GWh of non-billed electricity was due to technical losses, meaning that approximately 11 GWh of losses were non-technical. Other negative impacts associated with non-technical losses include the following:

• If power consumption in a region increases to a level where the average demand exceeds the rating of transformers or other electrical equipment, for example due to power theft, this can lead to power quality problems such as voltage collapse or transformer overloading which could result in load shedding. Frequent power cuts will lead to a deterioration in the customer–utility relationship, as consumers do not believe that the utility is capable of supplying the necessary service which could exacerbate the problem of non-technical losses as consumers refuse to pay for an ever-deteriorating service.
• Illegal connections to the electricity network often traverse roads, fields and footpaths and present a safety risk for communities who must go about their normal daily activities with the risk of coming into contact with live cables or wires that have not been properly installed.
• Electricity meters are usually co-located with fuses, circuit breakers, residual current earth leakage detection and other safely devices. Bypassing of a meter will often lead to unintentional bypassing of this protection, creating the risk that electrical faults or excessive demand within the house may not be detected. The supply will remain live within the house, increasing the risk of electrocution or fire. The demand on the local network may also exceed its capability, leading to disturbance to other local customers or damage to utility equipment.

3. Contributing Factors to Non-Technical Losses

3.1. Affordability

In many developing countries, the cost of a unit of electricity is substantial when compared to income. For example, more than 70% of Nigerians live on less than 3.20 US Dollars per day [20], yet the cost to power a small refrigerator for one year could account for up to 8% of this daily income [21,22,23]. When additional appliances such as phone chargers, light bulbs and cookers are accounted for, it is therefore not surprising that research shows there is a low willingness to pay for electricity in many developing countries, which is often driven by ability to pay, i.e., low levels of income, predictability of income flow and credit constraints [17]. However, inability to pay or low willingness to pay does not mean that residential consumers do not desire access to electricity. As such, it would appear logical to conclude that residential consumers with low levels of income are most likely to steal electricity, thereby contributing significantly to non-technical losses of electricity. However, from the literature review, it is difficult to come to such a conclusion as, rather than resorting to theft, many residential consumers will modify their behaviours to try and ameliorate the effects of high electricity costs. Tesfamichael reports that in Ethiopia, in response to increasing tariffs, many households changed their daily routines by reducing electricity consumption and/or increasing the use of biomass to curb increased electricity costs [24]. This behavioural change was also observed in Georgia [25], where between 1996 and 2002, during their energy crisis, fairly stable energy expenditure and consumption levels suggested that households in Tbilisi, in response to tariff increases, appeared to be replacing electricity with less expensive fuels. Furthermore, qualitative findings in Moldova [25] suggest that theft is not related to income and that wealthy customers were most likely to steal as they had the means to bribe meter readers, invest in technology to circumvent the meter, or steal using other means. While poor households were found to steal, it was often only occasionally or because they were disconnected for failure to pay their bills. Winther observed similar issues in Zanzibar [26], where she noted that it was the most privileged individuals who committed plain stealing and that their higher level of consumption, as well as their social and political power, may account for why they are more likely to exploit the system compared to those who are less privileged.

Furthermore, a distinction can be drawn between the affordability of using electricity and the affordability of access (related to the costs of connection) [27]. The initial cost of connecting to the electricity grid can be a barrier to connection as, for example, willingness to pay in Liberia fell from 90% to 60% when the connection charges moved from zero to USD 10 and fell further to about 10% when the proposed connection charges exceeded USD 50 [17]. High connection fees can lead consumers to seek alternative means to secure electricity including meter sharing. This involves the illegal connection of a house to the electricity grid through the meter of a legally connected neighbouring house. There will usually be an agreement between the parties as to how to pay the bill. This practice of sharing connections among several households is widespread in Africa, and for example, in Ethiopia, the number of households reporting connection to the grid exceeds the number of residential customers reported by the utility by 150% [6]. On the face of it, leaving aside the safety issues associated with meter tampering, from a financial perspective meter sharing would not appear to be a significant issue for utilities assuming that the sum of all power being consumed is metered correctly and paid for. However, the practice of meter sharing compounds the issues of unaffordability and energy poverty for many electricity consumers in developing countries. This is because governments in many developing countries provide the poorest consumers with either a free or heavily discounted initial allocation of electricity, often referred to as a “Lifeline Tariff”. For example, in 2003 the South African government introduced a policy where the first 50 kWh/month of electricity was free for all grid-connected “poor households” [28]. The practice of sharing electricity meters to offset grid connection charges means that the user, at whom the policy is targeted, will not be able to benefit from the subsidised tariff as the combined consumption of multiple households places them in higher-priced tariff brackets [6].

3.2. Customer–Utility Relationship

Alongside affordability challenges, it appears that citizens in developing countries may refuse to legally connect to the electricity grid as a form of protest. The World Bank [17] suggests that such protest is a direct result of poor service delivery by utility companies, which generates distrust among citizens in the ability and willingness of the utility to deliver sufficient value for money. As a result, households may resort to theft, which in turn leads to more frequent undervoltage events, thereby forcing the utility to disconnect customers more often to protect the electricity system, which exacerbates the feeling of distrust. This downward cycle of ever decreasing service quality, coupled with decreasing willingness to pay, leads to increased levels of non-technical losses on the system, which in turn adversely affects the financial performance of utility companies. As a result of reducing financial performance, utilities often must resort to some or all of the following measures:

• Increased electricity tariffs to cover the cost-of-service provision from a reduced compliant bill paying consumer base;
• A reduction in the quality-of-service provision such as deferred capital expenditure, a slower customer connection rate or delays to planned maintenance activities;
• A reduction in the funding of public services such as education or healthcare if government funds are diverted to stabilise the financial performance of the electricity utility.

Such measures can have severe impacts on consumers as evidenced in:

• Sierra Leone where, due to insufficient revenue, the overhead distribution network for the low-income eastern part of the town was cannibalised for spare parts to repair the network of the high-income western part of the town leaving many customers without power [29];
• Liberia where the CEO of LEC in an email (P Buckley 2021, personal communication, 20 August) confirmed that the detection and replacement of faulty meters usually took weeks or even months, and due to lack of electricity supply resulting from the faulty meter and the long-time taken to rectify the situation, customers often bypassed the meters thereby consuming electricity without paying for it;
• Latin America where state owned utilities were performing poorly and an ever-deteriorating service quality to customers created a vicious downward circle, or low-level trap, where the willingness of the population to pay for higher tariffs steadily declined over time, thus reducing the income source for the power sector and deepening the crisis [3].

The correlation between a poor customer–utility relationship and high non-technical losses is not limited to centrally planned electricity systems such as the ones referenced above but is also evident in decentralised systems such as the Sunderban Islands, West Bengal, India where the state-owned utility established a series of small solar photovoltaic power plants, incorporating battery banks and mini-grids, to provide the electricity. Winther [26] observed that for about a decade the rate of collected revenues on the Sunderban Islands was high, theft was almost non-existent, and defaults were few, but over time, due to lack of investment in the system, the physical quality and condition of many of the battery banks deteriorated and plant performance suffered. Furthermore, the utility company did not upgrade the system to address evolving customer needs, such as the requirement to charge multiple mobile phones, and this lack of flexibility contributed to the practice of overconsumption, thereby further reducing the quality of battery performance and supply. As a result of this reduction in performance and lack of investment by the utility, non-technical losses increased, and Winther contends that it was the supplier’s failure to deliver a service that customers paid for which finally jeopardised their trust and led to a lack of compliance. Furthermore, Winther concludes that the core customer–provider relationship is a key to understanding how unsustainable energy practices such as theft may be avoided.

3.3. Culture of Non-Payment

The literature suggests that a “culture” of non-payment [30] in certain countries can also lead to increased non-technical losses. Winther [26] expands on this concept that a person’s decision to steal from or defraud the electricity utility can be influenced by social norms, i.e., a “culture of non-payment”, and defines a scale for the local moralities and ethics involved with electricity theft in Zanzibar. For example in Zanzibar meter sharing is a socially accepted form of theft and Winther states that the reasons for this could be due to local customs, for example, husbands with more than one wife supplying electricity to several households [26]; as such, an accepted social norm has created a culture of non-payment. However general conclusions pertaining to “cultures of non-payment” are difficult to make, and from the literature, it is clear that each country has its own unique issues; as such, the particular circumstances in each country need to be analysed. For example:

• During the early 1990s in Armenia, common explanations for the non-payment of electricity included: the government owed salary arrears or pension arrears, savings disappeared from bank accounts at the end of 1993, and public sector employees often stated that until wages were increased, they would not pay their utility bills [25].
• There can be long delays associated with obtaining an official connection which leads people to seek alternative means of obtaining electricity. For example, it can take up to 450 days in Madagascar and 455 days in Guinea-Bissau to obtain a grid connection [31].
• In Liberia, according to an email from the CEO of LEC (P Buckley 2021, personal communication, 20 August), in a post-civil war post-Ebola virus environment, non-payment of electricity bills is seen as a minor transgression by many residential consumers who have had to endure and survive recent disasters.
• Many houses in developing countries are not built to a standard that facilitates connection to the electricity grid, and if the utility refuses to connect a household until the necessary works have been completed, then this can contribute to consumers seeking an informal connection.
• According to the World Bank, people who live in informal settlements (“slums”) often respond to poor service quality by ceasing to pay their bills or stealing electricity. When the distribution companies attempt to enforce revenue collection, residents and organized criminals often become hostile to their efforts, resulting in the utility company withdrawing totally from the area. Consequently, after years of receiving a service at no cost, a culture of non-payment becomes engrained where consumers come to believe that they have a right to free electricity [32].

3.4. Corruption and Political Interference

The World Bank has reported that when total losses are high, in excess of 25%, large consumers usually account for a large fraction of the losses. They refer to this as the Pareto effect, whereby a small percentage of customers account for a disproportionately high fraction of the total revenue [3]. In such circumstances, there is a strong economic incentive for corruption between utility company employees and these large consumers for whom electricity costs are a large portion of overall expenditure. For example, in Georgia during the late 1990s, corrupt employees diverted electricity supply to some non-paying public sector consumers which deprived the utility company of sufficient power for its paying customers, particularly during periods of high demand in winter [25].

In several Indian States, electricity consumers involved in agriculture are unmetered and are provided with electricity either free of charge or are required to pay a fixed amount for electricity irrespective of the amounts consumed. With the possibility of being voted out of office because of a vocal and well-organised agricultural sector who are opposed to electricity pricing reforms, politicians have little incentive to support such reforms and as a result, few, if any, meters are installed, prices are unchanged, and theft continues to plague the sector [33]. To take advantage of this situation, some consumers illegally tap electricity from locations on distribution feeders close to agricultural equipment, which further compounds the problem of non-technical losses on the system [5].

4. Combating Non-Technical Losses

4.1. Prepay and Tamper Proof Meters

Radical changes in payment regimes are taking place in Sub-Saharan Africa as utilities are moving away from a post-paid electricity system to a pre-paid system [30]. From the utility’s perspective, the ability to instantly disconnect consumers, once the pre-paid units run out, has the advantage of improving overall system reliability. As supply can be matched to demand from consumers who are paying for the power, the problem of collecting revenue from customers who have consumed electricity but not paid for it is no longer an issue and there should be increased operational efficiencies. Coupled with the installation of tamper proof meters, which according to a Moldovan utility company resulted in a 9.3% fall in theft between 1999 and 2002 [25], pre-paid meters can be an effective tool in the fight against electricity theft and non-technical losses.

However, Tesfamichael [24] and Jacome [30] contend that pre-payment schemes can have an adverse impact on lower income households who may only be able to pre-purchase small quantities of electricity which entrenches energy poverty in vulnerable households, and while the change from post-paid to pre-paid meters may relieve poor households from debt, such relief comes with the fear of being left in the dark. More importantly, where service quality is poor, pre-paid customers are particularly vulnerable, as they are being asked to pay in advance for electricity that may not be delivered when needed.

The cost of implementing a pre-paid meter scheme, including the accompanying tasks of upgrading information technology systems, procuring tamper proof meters and building the capacity of utility employees, is a significant issue that also needs to be considered. In an email the CEO of the Liberia Electricity Corporation (P Buckley 2021, personal communication, 20 August) confirmed that it was estimated that the utility company would need to invest around USD 17.5 million over a five-year period (2016–2021) in order to reduce total losses from 32% to 11%. Liberia’s gross domestic product (GDP) in 2015 was USD 3.2 billion [34], and such an investment would represent 0.6% of GDP so this is not a decision that can be taken lightly by either the government or utility company. As well as the cost of implementation, an analysis on the likely effectiveness of a pre-payment scheme needs to be considered. For example, an email from the CEO of the Liberia Electricity Corporation (P Buckley 2021, personal communication, 20 August) confirmed that following the introduction of a pre-payment scheme in Liberia in 2016, non-technical losses initially increased, as the utility company overestimated what the system was capable of and did not employ sufficient resources to deal with non-payment and illegal connections.

4.2. Reducing Initial Connection Charges

As previously mentioned, high initial charges associated with connecting to the grid can lead consumers to seek alternative means of connection either by connecting illegally to a supply point or via meter sharing. A method that could be used to discourage these practices would be to make the initial grid connection free, or low cost, and to minimise fixed costs (standing charges) as part of the tariff structure. These subsidies could be recovered by increasing the charge per unit of electricity consumed over a certain threshold such that the poorest households could still access electricity and benefit from any government schemes such as “lifeline tariffs”, while wealthier households would have to pay higher electricity charges but, by virtue of their higher income, may be able to offset some of the cost increases through the implementation of energy efficiency measures or by moderating use. This would demonstrate an element of fairness to the poorest consumers and would also allow the utility to engage with such consumers to explain why it is important to pay for electricity and the benefits associated with a reliable and stable electricity supply. However, when designing subsidy schemes, governments and utilities do need to ensure that cost-recovery will be achieved such that a reliable electricity system, providing the necessary level of service, can be maintained so as to avoid a deterioration in the customer–utility relationship, which in turn could lead to an increase in non-technical losses (as previously described). While there may be political or social sensitivities associated with increasing electricity charges for wealthier consumers, the World Bank contends that these should not be overemphasised, and where utility companies are operating with large deficits, or in countries where there are low access rates to electricity, then those who can afford higher tariffs could be asked to cross-subsidise low-income consumers, as long as the latter’s total consumption is a small fraction of the total electricity sold [6].

4.3. Strengthening the Law and Law Enforcement

The contribution of pre-paid electricity meters in reducing non-technical losses cannot be overlooked, but a key point made by Mwaura [35] is that the implementation of pre-paid electricity meters alone is not sufficient to reduce electricity theft and that other government policies such as increasing the penalties for those caught engaging in illegal power consumption are required. Research undertaken by Briseño and Rojas [36] into some of the factors associated with electricity theft in Mexico supports this point, as they conclude that it is important for government to send a signal to the public that stealing electricity has negative consequences for society and that the government can punish those engaged in such activities, which should reduce electricity theft. A successful example of this, as outlined by the World Bank [3], is in the Indian State of Andhra Pradesh where T&D losses, both technical and non-technical, during 1999 were estimated at 38%. Such losses were contributing to significant financial losses for the utility companies, and as a result, the government launched a comprehensive plan for controlling theft. Electricity theft was made a cognisable offense [37], a law was introduced which provided for mandatory imprisonment for offenders and collusion by utility staff was recognised as a criminal offense. Such measures, coupled with a number of other key changes within the electricity utility companies, yielded significant results as T&D losses decreased from 38% in 1999 to 26% in 2003, and the trend of loss reduction continued as for the fiscal year 2007/08, the distribution companies applied to the Regulatory Commission for a loss allowance of 18.7%.

4.4. Identify and Target the Worst Offenders

The World Bank reports that in Africa, unmetered consumption is disproportionately concentrated in those who can afford to pay cost-reflective tariffs, and revenue protection programs should first focus on large- and medium-size customers who usually account for the bulk of commercial and bill collection losses [6]. Such a situation was also observed in central and eastern Europe where those mainly responsible for non-payment of electricity bills were found in the corporate and municipal sectors, particularly among state owned enterprises [38]. However, policies to tackle non-payment usually centre around the “average residential user”, as evidenced in Zanzibar where the requirement for pre-paid meters and a “use only what you can afford policy” does not apply to large private sector companies, nor many government entities that are in heavy debt to the utility, while the public face of the utilities debt remains the lay citizen [30]. According to an email from the CEO of LEC (P Buckley 2021, personal communication, 20 August) in Liberia, large consumers account for approximately 60% of non-technical losses while representing less than 10% of the customer base, and to combat theft from large consumers LEChas taken measures such as:

• the installation of high security tamper resistant metering systems (HSMs);
• monitoring of the HSMs via alarms and cameras;
• moving industrial consumers to a pre-paid system;
• incentivising customers by offering lower tariffs to comply with the installation of HSMs and pre-payment arrangements.

4.5. Regulation and Privatisation

As well as increasing penalties for those caught stealing electricity, the regulatory regime will likely need strengthening in countries attempting to reduce high levels of non-technical losses. Smith [10] contends that the highest rates of electricity theft occur where the electricity sector is state owned and state managed because: state enterprises do not try and optimise profits in the way a “true” commercial enterprise would, they are intertwined into the political and bureaucratic structures, and there are few incentives to reduce theft. An analysis undertaken by the World Bank [3] would appear to support this contention, as a comparison of private sector and state-owned electricity distribution companies shows that private companies had higher labour productivity and operational efficiencies. However, Smith states that caution needs to be exercised about promoting privatisation as a panacea for the ills of inefficiency [10] and the World Bank also caution that any results showing increased performance from private sector utilities needs to be complemented with an analysis of the specifics in each case [3]. Furthermore, implementing a strong and independent regulatory environment may be difficult in many countries, as governments can pressure regulators to modify or overturn decisions and tariff-setting can be highly politicised, as governments are sensitive to popular resentment against price increases that are often necessary to cover costs [29].

4.6. Community Engagement and Other Measures

As concluded by Winther in the example from the Sunderban Islands [26], it was the supplier’s failure to deliver a service that customers paid for which finally jeopardised their trust and led to a lack of compliance. In an email (P Buckley 2021, personal communication, 20 August), the CEO of LEC advised that to prevent such issues arising in Liberia community leaders are engaged with the aim of effecting culture change such that communities feel they are part of the solution to tackling non-technical losses and poor service provision and ultimately see the benefit in paying for electricity. Community support is formalised via a Memorandum of Understanding between LEC and the consumer, where LEC on its side endeavours to respond promptly to complaints and notifications of transgressions. This “formalised contract” has been successful in improving the relationship between LEC and consumers as it gives the consumers a written performance metric should the utility need to be held to account, while at the same time setting out obligations on the consumer to which the utility company can refer if required. Other measures to reduce non-technical losses from residential consumers include:

• Moving the point of metering from the pole-top, where they are open to interference, to the consumers’ premises where greater ownership for the equipment is likely;
• Reinforcement and upgrading of the T&D network to improve the quality of supply to consumers as well as more regular inspections of meters to ensure they work correctly;
• The installation of ready-boards, pre-packaged electrical panels comprising plug points and sometimes a light bulb, which are an alternative to conventional house wiring and are a way of providing power to houses which may be “sub-standard” in terms of building quality.

5. Role of Renewable Energy

On the face, countries and regions that experience high levels of non-technical losses do not appear to also suffer a lack of impetus in their renewable energy programmes. For example, Brazil faces significant challenges associated with non-technical losses of electricity [18], yet its power generation mix is dominated by renewables. In the first ten months of 2020, hydropower, nuclear, wind, solar photovoltaic (solar PV) and biomass accounted for close to 90% of total generation, up 2% on the same period in 2019 [39]. However large-scale hydropower is the dominant power generating technology, accounting for approximately 75% of Brazil’s generation mix. This is not a technology that can be rapidly deployed, given the scale, environmental impacts and cost of developing these types of projects. Furthermore, given that large hydropower schemes are often located away from large population centres, this remoteness does not foster the sense of ownership that occurs when local communities can see the generating technology which gives a sense that they are part of the solution to non-technical loss reduction.

For countries experiencing significant levels of non-technical losses, coupled with challenges associated with lack of generating capacity and inadequate T&D infrastructure, even if the required conditions exist to exploit large-scale hydropower, alternative renewable technologies will be required in the short and medium term to ensure adequate supply of electricity to the population. Wind and solar PV are technologies that can be rapidly deployed, especially solar PV which can be deployed in both utility scale (grid connected solar farms in the megawatt to gigawatt scale) and residential rooftop scale (anywhere from 200 W up to say 2 kW in scale).

India, similar to Brazil, is a country that faces significant challenges with non-technical losses of electricity [33] and is a beneficiary of the continued drop in the cost of solar PV technology. During the four-year period between 2014 and 2018, the installed capacity of solar PV in India grew from 2.6 to 25 GW [40]. India is not the only beneficiary of falling solar PV costs as, according to the IEA [39], investment in power generation technologies in Africa during 2020 was dominated by renewables, including a tender for up to 80 MW of solar PV projects in Togo and the award of a 100 MW solar project in Tunisia.

According to BloombergNEF, in their 2020 New Energy Outlook report [41], trends in decentralisation, system integration, digitalisation and electrification will shape the future power grid, and the median power-plant size globally will drop as a result of the decommissioning of large thermal and nuclear plants and the adoption of wind and PV. As a result, decentralised solar PV has the potential to provide access to electricity to those who currently go without, can be deployed rapidly without the need for costly and long lead time rural electrification programmes, requires less technical expertise to operate and maintain compared to nuclear and/or large fossil fuel generating plant and has the capability to contribute to a reduction in non-technical losses by empowering communities through a sense of ownership and rebuilding the customer–utility relationship.

Baffi [18] suggests that in Brazil the introduction of distributed generation (DG), via rooftop solar PV into areas with high levels of non-technical losses, could be a more effective way of reducing these losses than the traditional method of searching for ways to block energy that flows illegally from the distribution grid. The concept being that illicit consumers would be transformed into regular prosumers (where the consumer of energy starts to also act as a producer), thereby reducing non-technical losses, improving the quality of service, reducing electricity bills for compliant consumers, reducing the maintenance costs of the distribution utilities and contributing to the improved socioeconomic situation of low-income communities. Winther’s research in the Sunderban Islands [26] gives an insight into the potential of decentralised power systems in areas that suffer from inadequate access to electricity as, initially when small solar PV plants were established (from 25 to 120 kWp), theft was almost non-existent and defaults were few, primarily because local community groups supervised electricity consumption within the villages and created awareness about electricity’s benefits to motivate people to pay for connections and consumption. Such a situation appears to corroborate Baffi’s suggestion [18] that making consumers into prosumers could reduce non-technical losses in certain sectors of society. However, there are challenges associated with the role out of DG schemes including:

• Even allowing for the rapid decrease in technology costs, especially solar PV, there is a large initial investment associated with the installation of rooftop solar PV. This is one of the main challenges associated with its adoption especially in developing countries that typically suffer from lack of access to electricity or experience significant amounts of non-technical losses. As such, government policies to provide finance or initial capital subsidies are necessary to promote the take-up of such schemes, but even if such policies are in place, lack of awareness of the potential of renewable energy among citizens in developing countries is a barrier to the deployment of such technologies. For example, in India there are government supports available to enable the deployment of rooftop solar PV technology, but because of lack of awareness about solar power systems, lack of education and poor training and development mechanisms people are not able to take the benefits of capital subsidies and other supports [42].
• The setting of adequate tariffs such that maintenance can be funded is a basic requirement. If adequate provision is not made for the maintenance and/or upgrading of DG systems (such as inverters and batteries in solar PV plants), then the reliability of the system will decrease, thereby straining the customer–utility relationship which could in-turn lead to an increase in non-technical losses, overconsumption, and in a worst-case scenario, a collapse of the electricity generating system, as was the case in the Sunderban Islands [26]. Striking the balance between access and affordability while at the same time ensuring that cost-recovery is achieved is a critical success factor.
• The design of electricity networks is typically based on centralised generation sources i.e., large power generating stations feeding power into the transmission system from where it was fed into distribution systems for supply to consumers. According to Passey [43], when penetration of DG rises above the network’s minimum threshold, more significant issues can arise in some networks, and changes to the network are required such as minimising VAR flows (reactive power flows), power factor correction, increased voltage regulation in the network and careful consideration of protections issues such as fault current levels and ground fault overvoltage issues. As a result, utilities may be hesitant to allow the widespread take-up of DG schemes until grid-connection standards and agreements are developed, and given the limited resources that are often available to utilities in developing countries, the development of such standards may not be prioritised, leading to a stagnation in development [43].
• Empowering the local community through the use of local workforce was referenced as being a factor in the success of the Brazilian Luz para Todos (Light for All) electrification programme [44]. However, developing competencies to enable local communities to undertake the installation, operation and maintenance on renewable energy systems takes time to build and is an ongoing requirement over the operational life of the asset. In India, due to limited institutional capacities for workforce training, a lack of available skilled and semi-skilled workers could jeopardise the achievement of the government’s solar rooftop target of 40 GW by 2022 [42]. Passey [43] highlights the need not only for the training of the local workforce in the operation and maintenance of DG installations but also the importance of post-installation assistance programmes to monitor the performance of systems and to deal with manufacturers in relation to warranty claims for the replacement of broken components. Many past and present aid-based projects have not taken this into account, resulting in failure of the installed energy generation system, or even worse, damage to equipment connected to the DG energy supply [43].

6. Conclusions

Non-technical loss of electricity often presents a major additional challenge to utility companies in developing countries that are already grappling with overstretched infrastructure and low levels of available investment. The lost revenue associated with non-technical loss places a significant financial burden on a utility, generally leading to higher tariffs for those customers who are paying and to a poorer service overall. Reduced maintenance of network assets and reduced capital expenditure lead to reduced quality-of-service, particularly power cuts, and longer wait times for customers seeking new connections. These effects lead to a downward spiral: as customers are unwilling to pay for a decreasing quality of service, non-technical losses increase, which reduces the income to utilities and results in a worsening of the situation. As well as the adverse economic impacts, unauthorised connections and meter tampering can lead to damage to equipment and are often unsafe.

It would be simplistic to assume that the poorest households are the main cause of non-technical loss; this study has shown that it is often wealthier customers, including state-owned enterprises and large businesses, that are responsible for the largest proportion of overall non-technical loss. Whilst customer poverty and lack of income are contributing factors, it appears that people who are genuinely unable to pay for electricity will tend to reduce their consumption or switch to alternative energy sources. Average residential customers constitute the majority of the customer base by number but may only account for a small portion of the non-technical electricity loss. As such, utility companies should focus efforts to reduce non-technical losses on large consumers and those who can afford to pay cost-reflective tariffs. Consistent enforcement is essential.

Replacement of credit meters with pre-paid meters can serve to reduce non-technical losses but can have negative impacts for vulnerable consumers who may only be able to purchase small amounts of electricity in advance and even then, may find that it is not delivered when needed due to supply interruptions.

The cost of the initial connection is a barrier for many people in developing countries, and the higher the charge, the less willing people are to pay for a legal connection. Modifying connection charges to incentivise legal connections among poorer communities, possibly by increasing the variable tariff to people who consume more power, could be a mechanism to reduce the number of illegal connections in some countries.

Improving the regulatory regime and facilitating private sector participation in the electricity sector are tools that can be considered by governments who want to tackle high levels of non-technical losses. However, incentivising private sector participation by ensuring tariffs are appropriate to cover costs and return a profit as well as ensuring that the electricity regulator is independent and free from political interference are challenges that need to be overcome.

Renewable energy, especially through distributed generation such as rooftop solar PV, has a role to play in both improving access to electricity and reducing non-technical losses in developing countries, but there are challenges associated with this, including: the initial capital cost of such systems, ensuring the systems are maintained properly, incorporating distributed generation into weak electricity grids or grids that have not been designed to accommodate such schemes, and the amount of time it takes to build competency and train people in such technologies. However, if local communities are consulted and involved in decisions around electricity provision and employment opportunities are provided for the local workforce, then renewable energy has a role to play in reducing non-technical losses of electricity.