Mobile money and multidimensional energy poverty: a cross-national study of Burkina Faso and Togo

Compaore, Eugène Dimaviya; Guira, Asmo; Maiga, Boukaré Article Mobile money and multidimensional energy poverty: a cross-national study of Burkina Faso and Togo Cogent Economics & Finance Provided in Cooperation with: Taylor & Francis Group Suggested Citation: Compaore, Eugène Dimaviya; Guira, Asmo; Maiga, Boukaré (2024) : Mobile money and multidimensional energy poverty: a cross-national study of Burkina Faso and Togo, Cogent Economics & Finance, ISSN 2332-2039, Taylor & Francis, Abingdon, Vol. 12, Iss. 1, pp. 1-24, https://doi.org/10.1080/23322039.2024.2399758 This Version is available at: https://hdl.handle.net/10419/321589 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Cogent Economics & Finance ISSN: 2332-2039 (Online) Journal homepage: www.tandfonline.com/journals/oaef20 Mobile money and multidimensional energy poverty: a cross-national study of Burkina Faso and Togo Eugène Dimaviya Compaore, Asmo Guira & Boukaré Maiga To cite this article: Eugène Dimaviya Compaore, Asmo Guira & Boukaré Maiga (2024) Mobile money and multidimensional energy poverty: a cross-national study of Burkina Faso and Togo, Cogent Economics & Finance, 12:1, 2399758, DOI: 10.1080/23322039.2024.2399758 To link to this article: https://doi.org/10.1080/23322039.2024.2399758 © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group Published online: 06 Sep 2024. Submit your article to this journal Article views: 907 View related articles View Crossmark data Citing articles: 1 View citing articles Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=oaef20 DEVELOPMENT ECONOMICS | RESEARCH ARTICLE Mobile money and multidimensional energy poverty: a cross-national study of Burkina Faso and Togo Eugène Dimaviya Compaore, Asmo Guira and Boukar e Maiga Center for Economic and Social Studies, Documentation and Research (CEDRES), Thomas SANKARA University, Ouagadougou, Burkina Faso ABSTRACT Using data from the FinScop survey (2016), this study aims to analyse the effect of mobile money on multidimensional energy poverty (MEPI) in Burkina Faso and Togo. MEPI was calculated using the method of Alkire and Foster in 2011, and a linear regression based on the instrumental variable strategy was applied to control for endogeneity bias arising from the dual causality between energy poverty and mobile money. To test the robustness of our results, we used the endogenous switching regression (ESR) model to resolve the problems of self-selection bias and endogeneity. The average effects of treatment on the treated (ATT) and the untreated (ATU) were calculated from the coefficients of the ESR models. The incidence of multidimensional energy poverty was estimated at 90.7 and 91.1% in Burkina Faso and Togo, respectively. In addition, we found substantial differences between the subgroups in terms of multidimensional energy poverty in each of the countries. The results also robustly indicate that an increase in mobile money adoption per standard deviation is associated with a reduction in multidimensional energy poverty of −0.402 standard deviations in the case of Burkina Faso and −0.628 standard deviations in the case of Togo. We argue that mobile money can be an effective policy tool in the fight against energy poverty in developing countries. IMPACT STATEMENT This research explores the impact of mobile money on multidimensional household fuel poverty by focusing on two developing countries (Burkina Faso and Togo). The comparative study between these two countries aims to shed light on the mechanisms by which mobile money influences household energy poverty. By analysing the effects of mobile money in different contexts, the study seeks to establish whether the results observed can be generalised. If the impact is similar in the two countries despite their differences, this would allow the conclusions to be applied on a more global level. The results highlight the need to integrate mobile money into policies aimed at reducing household energy poverty in developing countries. ARTICLE HISTORY Received 6 March 2024 Revised 4 July 2024 Accepted 23 August 2024 KEYWORDS Burkina Faso; instrumental variable; mobile money; multidimensional energy poverty; Togo SUBJECTS African Studies; Economics; Finance; Political Economy; Economics and Development JEL CODES D14; O12; D12; I32; O55 1. Introduction Access to reliable, sustainable and modern energy services at an affordable cost is essential for socioeconomic development (Koomson & Danquah, 2021; Nussbaumer et al., 2012). It helps to reduce poverty by enabling other activities that are vital for human development (Birol, 2007; Bridge et al., 2016). Indeed, this access facilitates the use of certain technologies that are likely to reduce poverty through the channels of education, labour productivity and health (Bridge et al., 2016; Sambodo & Novandra, 2019). Also, the use of clean energy ensures a decent life for households because of its usefulness for cooking, lighting and commercial activities (Phoumin & Kimura, 2019; Reilly, 2015). However, in Sub-Saharan Africa (SSA), almost 53% of the population has no access to electricity, and almost 85% of the population relies on inefficient cooking systems (International Energy Agency (IEA) CONTACT Eugène Dimaviya Compaore [email protected] Center for Economic and Social Studies, Documentation and Research (CEDRES), Thomas SANKARA University, 12P 417 Ouagadougou, Burkina Faso. ß2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. COGENT ECONOMICS & FINANCE 2024, VOL. 12, NO. 1, 2399758 https://doi.org/10.1080/23322039.2024.2399758 et al., 2020). People’s reliance on inadequate energy services is a form of poverty as it limits their ability to meet their energy needs and realise their full potential (Practical Action, 2014). Energy poverty is defined as "the lack of sufficient choice to access affordable, reliable, high quality, secure, and environmentally sound energy services to support economic and human development" (Reddy et al., 2010). Limited access to clean fuels (electricity and gas) and their affordability are two factors that contribute to this energy poverty (Hosan et al., 2023). Low household access to modern energy services can be explained by low purchasing power (Koomson & Danquah, 2021). Indeed, financial inaccessibility can force a person to modify their energy consumption in terms of quantity; this affects their standard of living, putting them in a situation of energy poverty (Lewis, 1982). Access to financial services can serve as a catalyst for improving household living conditions by enhancing their financial autonomy in the face of potential shocks (Afawubo et al., 2020; Duflo, 2020; Zhang & Posso, 2019). Previous studies have shown that financial inclusion plays a crucial role in poverty reduction and household well-being in general (Bukari & Koomson, 2020; DablaNorris et al., 2021; Koomsom et al., 2020). Koomson and Danquah (2021) show that financial inclusion positively affects energy poverty reduction. These earlier studies focused on formal financial inclusion. Given that in developing countries, a large part of the population is excluded from using formal financial services (Apeti, 2022) it is appropriate to explore other alternatives that could help in alleviating energy poverty in these countries. In this regard, the development of mobile telephones can provide an alternative by allowing poor households to access basic financial services via mobile money (Apeti, 2022). Advances in new technologies have facilitated the removal of certain barriers to accessing financial services through mobile money. These services via mobile phones promote the financial inclusion of vulnerable groups by reducing transaction costs (Bachas et al., 2018), access distances to financial services (Ahmad et al., 2020) and information asymmetries (Ahmad et al., 2020; Aron, 2018). Mobile money can also reduce poverty by transferring funds to the poor. It lowers transaction costs, and access to mobile money improves household welfare through economic security, increasing consumer spending (Haushofer & Shapiro, 2016; Hjelm et al., 2017; Munyegera & Matsumoto, 2016). Furthermore, it aids poverty reduction by improving household resilience to shocks. Mobile money increases people’s propensity to save and, therefore, provides easy and rapid access to financial resources, especially for the most disadvantaged, thereby improving their ability to deal with health emergencies and react to negative shocks (Afawubo et al., 2020; Ky et al., 2018). Hence, access to financial services via mobile money is an effective approach to financially alleviating poverty, including that of vulnerable people ( AlvarezGamboa et al., 2021; Koomson et al., 2020; Polloni-Silva et al., 2021). The literature indicates that financial inclusion affects energy poverty through other transmission channels (Figure 1). According to Koomson and Danquah (2021), financial inclusion affects fuel poverty through its ability to increase household income and reduce monetary poverty and inequality, on the one hand, and through its investment effects in education, health and employment on the other. Zang et al. (2023) argue that the key factors determining fuel poverty are the price of energy, household income, access to energy appliances, a household’s specific energy needs, lower levels of education and geographical barriers. Wachtel (2001) identifies the promotion of higher savings rates, through the provision of attractive and innovative products to encourage savings mobilisation, as one of the channels through which financial development influences a country’s economic development. In addition, the use of mobile telephones affects energy poverty in economic and social terms (Zang et al., 2023). From an economic perspective, the use of telephones can reduce the problems of information asymmetries in a households’activities, while improving their bargaining power and hence their profitability, which improves their well-being (Jensen, 2007; Quandt et al., 2020). Moreover, it can enhance family well-being through increased consumption (Labonne & Chase, 2009). Oezcan et al. (2013) found that overall family well-being had a significant impact on energy choice. On a social level, the use of mobile phones diminishes fuel poverty by influencing households’social capital. Indeed, the use of the mobile phone as a means of acquiring and exchanging information also affects trust with close friends, family or strangers in the social circle (Kushlev & Proulx, 2016). While Churchill and Smyth (2020) argue that trust stimulates the dissemination of effective information over social networks and that social trust is closely related to income poverty, productivity and the provision of public goods, opportunities in the labour market all play an important role in determining the degree of fuel poverty. 2 E. D. COMPAORE ET AL. Therefore, the use of mobile phones can affect social capital, which plays a mediating role in the impact of mobile phone use on fuel poverty (Zang et al., 2023). Empirical studies indicate that mobile money has a positive impact on the probability of saving in mobile money savings accounts and receiving remittances on the consumption expenditure, on the poverty reduction and the household (Churchill & Marisetty, 2019; Djahini-Afawoubo et al., 2023; Islam et al., 2022;Naito et al., 2021; Suri & Jack, 2016). As for the empirical literature on the effect of mobile money on energy poverty, it remains little explored. Nevertheless, Zang et al. (2023) show that mobile phone use has a positive impact on reducing energy poverty in Tibet and China. According to these authors, mobile telephones reduces energy poverty by increasing household income or by increasing their social capital. Koomson and Danquah (2021) find that an increase in financial inclusion is associated with a reduction in household energy poverty in Ghana. Similar findings were reported by Dogan et al. (2021)inTurkey,Wangetal.(2023) in China and Sen et al. (2023) in Bangladesh. However, an analysis of these studies reveals that they have not explicitly examined the effects of mobile money alone on energy poverty. Furthermore, its effect on multidimensional energy poverty has not been sufficiently explored in developing countries. Given the findings that financial inclusion improves household welfare through its ability to mobilise savings to increase income, reduce poverty, and inequality, and thereby facilitate investment in education, health and family businesses, as well as efficiency gains, it is imperative to analyse the relationship between financial institutions and energy poverty. Therefore, this article aims to assess the capacity of financial inclusion through mobile money to reduce multidimensional household energy poverty in developing countries. This research contributes to the literature in three ways. Firstly, it departs from previous studies examining the relationship between financial inclusion and fuel poverty, specifically examining the relationship between mobile money and multidimensional fuel poverty and providing strong evidence for two developing countries with high rates of multidimensional fuel poverty. In addition, the study takes a multidimensional approach to analysing the link between Figure 1. Mobile money transmission mechanism for reducing energy poverty. Sources: authors based on literature review. COGENT ECONOMICS & FINANCE 3 mobile money and energy poverty. Energy poverty is a complex issue that lacks a consistent definition and is often difficult to quantify (Hosan et al., 2023). Furthermore, in developing countries, household energy poverty may involve simultaneous deprivations in terms of availability and affordability. It is, therefore, essential to measure energy poverty using a composite indicator that captures simultaneous affordability and availability deprivations (Crentsil et al., 2019; Hosan et al., 2023), which can be used to monitor progress towards the Sustainable Development Goals (Goal 7). This study is particularly important for developing countries, as it will help policymakers to formulate appropriate policies to reduce household dependence on traditional and polluting energy services. Secondly, the article takes into account the risk of endogeneity associated with selection bias and the dual causality between mobile money and fuel poverty. In this case, the variable of interest, mobile money take-up, may be endogenous (Wooldridge, 2002). To resolve the problems of endogeneity and selection bias, two methods are used: the instrumental variable strategy and the endogenous switching regression model. Indeed, Nan et al. (2021) pointed out that the majority of studies on the socioeconomic effects of mobile money have not used rigorous methodologies. Therefore, they suggested that future research should adopt more rigorous techniques, such as the instrumental variable technique, to control for endogeneity bias in order to obtain more robust results, and such a suggestion is considered within the scope of this article. Using data from the FinScop survey (2016) conducted in Burkina Faso and Togo, this study aims to contribute to the growing literature on the determinants of energy poverty in developing countries. It is one of the few studies that examines the link between mobile money and energy poverty by providing robust evidence for two developing countries. In addition, the analysis is conducted as a comparative study between the two countries to provide insight into the mechanisms by which mobile money affects household fuel poverty. This comparative approach enhances our understanding of the causal relationship between mobile money and household well-being. Therefore, if the study reveals that the effect of mobile money on energy poverty reduction is similar in both countries, despite their differences, it suggests that the conclusions drawn can be applied globally. However, if disparities are observed, they would need to be applied on a case-by-case basis. This article aims to analyse the effect of mobile money on energy poverty in Burkina Faso and Togo using a multidimensional measure of energy. The study covers a Sahelian country and a coastal country, i.e. Burkina Faso and Togo, respectively. Although the level of access to electricity remains a matter of some concern in these two countries, part of Burkina Faso’s electricity demand is met by imports from Togo. Approximately 86% of households in Burkina Faso and 49% of households in Togo do not have access to electricity (International Energy Agency (IEA) et al., 2020 1 ). For most households with access to electricity, reliability of supply and affordability remain significant concerns. Additionally, we note that almost 90% of households in Burkina Faso and almost 91% of households in Togo do not have access to clean cooking fuels and technologies (International Energy Agency (IEA) et al., 2020). This shows that firewood and charcoal are the main cooking fuels for 91.8% of these households in Burkina Faso and 49.3% in Togo (INSD, 2021; 2 INSEED, 2020) 3 . Indeed, fuel availability, lack of clean energy sources and their availability are the main reasons for the high use of solid fuels in developing countries (Khanna et al., 2019). Moreover, bank penetration rates are low in these countries, at 20.6% in Burkina Faso and 27% in Togo, according to the BCEAO 4 (BCEAO, 2021). However, with the emergence of mobile money services, the rate of use of financial services is increasingly compensated, leading to rates of 70.9% in Burkina Faso and 72.2% in Togo. In this context, mobile money is likely to assist households to meet their basic energy needs in these two countries, thereby helping to reduce energy poverty and improve social welfare. Consequently, this study aims to investigate the impact of mobile money on household energy poverty in these countries where financial inclusion is strongly enhanced by mobile phones. Given that mobile money has an impact on a household’s ability to mobilise savings, it may have an impact on fuel poverty. Moreover, energy poverty can have a considerable impact on a household’s well-being since energy is an integral part of daily life (Hosan et al., 2023). The rest of the study is organised as follows. Section 3 presents various measures of fuel poverty, mobile money transmission channels and the relationship between mobile money and fuel poverty, while Section 4 presents the methodology, and describes the data and variables used. Section 5 analyses and discusses the results, and the final section presents the conclusion and policy implications. 4 E. D. COMPAORE ET AL. 2. Literature review 2.1. Measuring energy poverty In the literature, no clear and universal definition of energy poverty exists (Sadath & Acharya, 2017). According to these authors, domestic energy deprivation, whether due to a lack of heating fuel in developed countries or a lack of access to electricity in developing countries, signifies energy poverty and has similar socio-economic consequences on society’s well-being. However, Lewis (1982) defined energy poverty in terms of financial inaccessibility, whereby inadequate energy usage results from low-income levels, impacting people’s standard of living. Regarding income, Leach (1992) showed that the ratio of energy consumption to household income is higher in low-income households than in middle-income households. This led to the introduction of the 10% income threshold by Boardman (1991). Based on this threshold, a person whose energy consumption exceeds 10% of their income is classified as experiencing energy poverty. These definitions of energy poverty are mainly linked to affordability issues, whereas the situation appears to be more complex in developing countries. Energy poverty is perceived as a singledimensional issue, primarily linked to income, in developed countries. However, in the analysis of the relationship between energy poverty and poverty, there are other important factors than income, such as the price or cost of energy in the assessment of energy poverty (Barnes et al., 2011; Hills, 2012). This illustrates that energy poverty integrates multiple dimensions in the same logic as Amartya Sen’s capability approach to development and goes beyond the income or expenditure incurred for energy consumption (Pereira et al., 2011; Sadath & Acharya, 2017). In this context, energy poverty represents an inability to realise essential capabilities due to insufficient access to affordable, reliable and secure energy services, considering alternative means to attain these capabilities in a reasonable manner (Day et al., 2016). In addition, households cannot adequately heat their homes or meet other essential domestic energy services at affordable costs (Pye et al., 2015). Access to clean, modern, and affordable energy is recognised as facilitating the attainment of several Sustainable Development Goals (SDGs), including Poverty eradication (SDG1), Hunger eradication (SDG2), Good health and well-being (SDG3), Quality education (SDG4), Gender equality (SDG5), Decent work and economic growth (SDG8), Industry, innovation, and infrastructure (SDG9) and Sustainable cities and communities (SDG11), as outlined by Sambodo and Novandra (2019) Thus, recent studies have considered energy poverty from a multidimensional perspective (Day et al., 2016; Ogwumike & Ozughalu, 2016; Sadath & Acharya, 2017). Within this multidimensional framework, Nussbaumer et al. (2012) propose a multidimensional energy poverty index that considers factors such as energy access and other energy-related development considerations, including the type of cooking fuel and ownership of electrical appliances, e.g. refrigerators, radios, televisions and mobile phones. This research follows the same logic and uses this index for further analysis. 2.2. Theoretical literature: mobile money transmission channels Access to clean, affordable, stable, sustainable and modern energy sources improves socio-economic development and enables the achievement of the Sustainable Development Goals (SDGs), poverty eradication, gender equality, zero hunger, climate action, good health, and well-being (Crentsil et al., 2019; Rosenthal et al., 2018). The literature distinguishes between two main channels by which financial inclusion through mobile money can reduce fuel poverty. Mobile money contributes to the alleviation of energy poverty by reducing transaction costs and promoting financial inclusion across all segments of society (Aker et al., 2016; Alvarez-Gamboa et al., 2021; Koomson et al., 2020). Financial inclusion through mobile money improves household income and reduces inequality and poverty because it allows households to reduce economic shocks, maintain consumption levels over time and make future investments (Demirg€ uc¸-Kunt et al., 2017; Koomson & Danquah, 2021). To this end, Koomson et al. (2020) show that an increase in financial inclusion correlates with a decrease in a household’s probability of poverty and inhibits its risk of poverty. In this case, it offers blocked development prospects for disadvantaged segments of the population and reduces income inequality (Omar & Inaba, 2020; Park & Mercado, 2018). Promoting equal opportunities to access COGENT ECONOMICS & FINANCE 5 financial services fosters the economic integration of socially excluded households and maximises society’s overall well-being (Nanziri, 2016; Sani Ibrahim et al., 2019). This means that an increase in inequality or poverty has increasing effects on fuel poverty, whereas an increase in income reduces fuel poverty (Bouzarovski & Petrova, 2015; Casillas & Kammen, 2010; Churchill & Smyth, 2020; Khandker et al., 2010; Sadath & Acharya, 2017). This underscores the potential of financial inclusion to influence fuel poverty by impacting household income, poverty levels and inequality. Financial inclusion can also reduce energy poverty through its ability to mobilise savings (Afawubo et al., 2020; Ky et al., 2018) needed for improvements in education, health and work (Matekenya et al., 2020; Njiru & Letema, 2018; Sarma & Pais, 2011; Stein & Yannelis, 2019). These improvements can increase the likelihood of higher earned incomes and, thus, increase households’means of acquiring reliable, stable and sustainable energy sources for basic needs. Indeed, increasing the level of education reduces fuel poverty, although this effect is not significant in some studies (Crentsil et al., 2019; Sadath & Acharya, 2017; Sharma, 2016). Moreover, fuel poverty decreases as the education level of the household head increases (Crentsil et al., 2019). This implies that the level of energy poverty decreases for households with a high level of education. Financial inclusion makes it possible to earn income for immediate needs, such as health care, in order to maintain or improve the level of productivity in activities, which allows access to the necessary energy sources. Conversely, the absence of these energy sources can considerably worsen health problems (Biermann, 2016; Kahouli, 2020; Llorca et al., 2020). Based on the existing literature, the mobile money transmission mechanism that forms the theoretical framework for this study can be represented in Figure 1. 3. Methodology 3.1. Measuring the multidimensional energy poverty index (MEPI) For the measurement of MEPI, the study adopts the framework of Nussbaumer et al. (2012) derived from the literature on multidimensional poverty measures including the Oxford Poverty and Human Development Initiative (OPHI) developed by Alkire and Foster (2011). The MEPI assesses the sources and types of energy in which the households are poor. It involves the first stage of identifying energy-poor households based on the range of energy deprivations suffered by households, and a second stage that aggregates the information to reflect energy poverty in a holistic way. The unit of analysis for MEPI is the household. It is an index of four dimensions, each comprising an indicator representing access to, and use of, basic energy services (Table 1). Each indicator is coded 1 if the household is deprived of the indicator and 0 otherwise. An unequal weighting was assigned to each indicator according to its importance in measuring multidimensional energy poverty. Consequently, a multidimensional energy deprivation score is determined for each household by adding the weighted deprivations (ci) in such a way that the deprivation score is between 0 and 1. The higher the deprivation experienced by the household, the higher the deprivation score. The maximum score was 1 if the individual suffered deprivation in all indicators. Formally we have: ciðkÞ¼w1I1þw2I2þ:::wdId(1) where ciðkÞrepresents the energy deprivation scores, Ii¼1 if the household is deprived of indicator i and Ii¼0 otherwise, and wiis the weight associated with indicator iwith Pd i¼1wi¼1: Table 1. Dimension, indicators, deprivation thresholds and weightings. Dimensions/Indicators Deprivation cut-offs Weights Lighting Electricity The household has no electricity 0.3 Cooking Modern cooking fuel The household cooks with biomass (firewood, cow dung, wood, charcoal) 0.3 Services provided by means of a household appliance Asset ownership The household does not own a refrigerator or freezer 0.2 Education and entertainment Education and entertainment The household does not own a radio or television 0.2 Source: Based on Nussbaumer et al. (2012). 6 E. D. COMPAORE ET AL. These deprivation scores are subsequently compared by a predefined poverty line ðkÞso that a household is considered multidimensionally poor if cik:Following Nussbaumer et al. (2012), we define a household as multidimensionally energy-poor if its deprivation score is greater than, or equal to, 0.33 ðk¼0:33Þ: The MEPI represents the product of the incidence and intensity of multidimensional energy poverty. The incidence, which represents the proportion of people considered to be energy-poor, was calculated as follows: H¼q n(2) where qis the number of multidimensionally energy-poor households and nis the total number of households. Energy poverty intensity indicates the average proportion of indicators for which multi-dimensionally energy-poor households are deprived. Poverty intensity ðAÞis the average deprivation score of households identified as energy-poor, and is obtained as follows: A¼Pn i¼1ciðkÞ q(3) The MEPI is obtained by multiplying the incidence and intensity of energy poverty: MEPI ¼HA¼Pn i¼1ciðkÞ n The MEPI respects the condition of dimensional monotonicity. It can also be broken down into subgroups (according to gender, place of residence, etc.). 3.2. Dimensions,indicators and thresholds of deprivation The dimensions, indicators and deprivation thresholds, as well as the weights used in the MEPI estimation for Burkina Faso and Togo, are presented in Table 1.Inthisstudy,fourdimensionsandindicatorswereused to measure multidimensional energy poverty. The choice of these indicators and deprivation thresholds is a relevant reflection of the household energy demand services. The weights are applied unevenly to reflect therelativeimportanceofmultidimensionalenergypoverty(Nussbaumeretal.,2012). 3.3. Specification and empirical methods According to the literature, the mobile money variable is considered to be potentially endogenous (Djahini-Afawoubo et al., 2023; Islam et al., 2022; Riley, 2018). The source of endogeneity may arise from the bidirectional causality inherent in the relationship between mobile money and multidimensional energy poverty. Indeed, we argue that mobile money can help to reduce household energy poverty. However, poor households seeking access to modern energy sources may choose to save via mobile money to achieve this (Koomson & Danquah, 2021). Thus, the decision to adopt mobile money may be influenced by the household’s decision to access modern energy services. Furthermore, the decision to adopt mobile money may be correlated with observable, as well as unobservable, characteristics, thus introducing bias due to the heterogeneity of individuals in the sample. We use an instrumental variable method to deal with this endogeneity problem (Bezu et al., 2014; Djahini-Afawoubo et al., 2023; Verkaart et al., 2017). This work uses a linear regression model with an instrumental variable (IV). The objective is to find an instrumental variable that is strongly correlated with moving money and uncorrelated with the error term in structural Eq. (1). Thus, we specify the model as follows: AM i¼dþbDistiþei1(1) EnPovi¼aþAM iþcXiþei2(2) where € EnPov€is the dependent variable representing multidimensional energy poverty. It is measured using household energy deprivation scores (see Eq. 1). € AM€is the adoption of mobile money. It is a COGENT ECONOMICS & FINANCE 7 Table 7. Mobile money and energy poverty (2SLS regression). Variables Burkina Faso Togo Mobile money adoption −0.2138 −0.3373 (0.0162) (0.0816) [−0.4015] [−0.6275] Age −0.0030 0.0015 (0.0008) (0.0015) [−0.1819] [0.0966] Age squared 0.0000 −0.0000 (0.0000) (0.0000) [0.1811] [−0.0807] Sex (female) −0.0163 −0.0145 (0.0055) (0.0104) [−0.0342] [−0.0303] Married −0.0072 −0.0089 (0.0062) (0.0077) [−0.0146] [−0.0183] Literacy −0.0510 −0.0065 (0.0076) (0.0102) [−0.1008] [−0.0093] Household size 0.0030 −0.0021 (0.0006) (0.0035) [0.0722] [−0.0320] Household size squared −0.0000 0.0001 (0.0000) (0.0002) [−0.0345] [0.0285] Area of residence (urban) −0.2523 −0.1588 (0.0111) (0.0161) [−0.4325] [−0.3284] Social capital 0.0117−0.0158 (0.0062) (0.0098) [0.0191] [−0.0332] Self-employed 0.0296 −0.0031 (0.0058) (0.0103) [0.0621] [−0.0044] Employed −0.1379 −0.0084 (0.0251) (0.0244) [−0.0748] [−0.0076] Constant 0.9106 0.8276 Age (0.0160) (0.0225) First-Stage Instrument (Time to reach the nearest mobile money agent) −0.0037 −0.0021 (0.0001) (0.0002) F-statistic of first regression 612.356 108.104 Observations 5013 5045 R-squared 0.4860 0.1000 Endogeneity test Durbin (score) chi2 (1) 73.2022 11.187 Prob 0.0000 0.0008 Wu-Hausman F (1, 4999) /F(1, 5031) 76.5231 11.366 Prob 0.0000 0.0008 Test of weakness of instruments R-sq 0.3422 0.1159 Adjusted R-sq. 0.3406 0.1138 Partial R-sq 0.1504 0.0092 F (1, 5000) / F (1,5032) 612.356 108.104 Prob >F 0.0000 0.0000 Weak-instrument-robust inference Anderson-Rubin Wald test F (1,5000) / F (1,5032) 189.33 17.81 P-val 0.000 0.000 Anderson-Rubin Wald test Chi-sq (1) / Chi-sq (1) 189.83 17.85 P-val 0.000 0.000 Stock-Wright LM S statistic Chi-sq (1) 180.20 17.38 p-val 0.000 0.000 Robust standard errors in parentheses p<0.01, p<0.05, p<0.1. Standardised coefficients are in square brackets. Source: Authors’computation. 14 E. D. COMPAORE ET AL. and more secure, mobile money can stimulate small-scale trade (Jack & Suri, 2014), which can contribute to improving household income. As a secure means of storing value, mobile money promotes household savings, enabling households to smooth their consumption of modern and cleaner cooking fuels. In addition, mobile money can help to increase household electrification rates through simpler and faster payment of electricity bills (Coulibaly, 2023). Our results are consistent with those of previous studies that have highlighted poverty reduction through the adoption of mobile money (Djahini-Afawoubo et al., 2023; Lee et al., 2021; Suri & Jack, 2016; Yao et al., 2023). Furthermore, access to basic financial services (mobile money) remains an opportunity to improve household well-being and to reduce poverty (Apeti, 2022). In developing countries, where a large proportion of the population is excluded from formal financial services, the development of mobile telephones offers an alternative, and an opportunity, for poor households to access basic financial services via mobile money (Apeti, 2022; Coulibaly, 2023). This enables them to increase their savings rate, and consequently the opportunities for financial autonomy (Duflo, 2020), 8 in order to cope with unforeseen short-term shocks (Demirg€ uc¸-Kunt et al., 2018) that may concern the purchase of electricity or gas for cooking. Urban residence had a statistically significant effect on multidimensional energy poverty. There is a decrease in multidimensional energy poverty of −0.433 standard deviations for urban residents compared to rural residents in Burkina Faso, and a decrease of −0.328 standard deviations for Togo. This result is consistent with those reported in the literature (Djahini-Afawoubo et al., 2023; Oum, 2019). Age has a positive effect on reducing energy poverty, while age squared has a negative effect on it. An increase in age per standard deviation is associated with a decrease in energy poverty of −0.182 standard deviations, whereas an increase in age squared per standard deviation is associated with an increase in energy poverty of 0.181 standard deviations, resulting in a U-shaped relationship between age and multidimensional energy poverty. Thus, there is a given age at which the relationship between multidimensional energy poverty and age becomes positive. Literacy has a positive effect on energy poverty reduction. The results indicate a reduction in energy poverty of around −0.101 standard deviations for literate individuals compared to non-literate individuals. Literacy enables individuals to understand the importance and adoption of energy-efficient practices better. Indeed, literate people can read and assimilate information on how to reduce their energy consumption and adopt more efficient technologies. This can help to reduce energy poverty. DjahiniAfawoubo et al. (2023), Mdluli and Dunga (2021), and Oum (2019) also find that education positively affects poverty reduction. The results also indicate a U-shaped relationship between household size and multidimensional energy poverty. In other words, for a small household, multidimensional energy poverty increases, whereas for large households, multidimensional energy poverty decreases as household size increases. This result corroborates with the findings of Koomson and Danquah (2021). Multidimensional poverty increases for small households owing to factors such as fixed costs remaining constant and higher proportional expenditure. Social capital has a positive effect on multidimensional energy poverty. Our results indicate an increase in multidimensional energy poverty of 0.019 standard deviations for individuals with social capital compared with those without social capital. This result, which seems paradoxical, can be explained by the level of the quality of social capital. Communities with high levels of social capital are able to work better together to implement sustainable energy solutions. This reduced their vulnerability to energy poverty. Djahini-Afawoubo et al. (2023) found a non-significant negative effect of social capital on multidimensional poverty. Multidimensional energy poverty decreases by −0.075 standard deviations when the individual is an employee, and increases by 0.062 standard deviations for self-employed individuals. Similar results have been reported by Churchill et al. (2020). Employment allows individuals to have a stable income to cover their basic energy needs, such as electricity and transport. Access to employment can also facilitate investments in energy-efficient systems. This contributes to the efficient use of resources; however, selfemployed workers tend to have irregular incomes because of the nature of their work, which can make it difficult to cover energy costs that are usually constant. COGENT ECONOMICS & FINANCE 15 4.8. Sensitivity/robustness checks We conducted several sensitivity checks to assess the robustness of the estimates. Table 8 presents the results of the ESR model estimates. In each country, column (1) presents the estimates of the factors influencing mobile money adoption, while columns (2) and (3) present the effects of mobile money adoption on multidimensional energy deprivation scores. The test of independence between the three equations, i.e., the likelihood ratio test in each country, is rejected in favour of a joint dependence of the error terms of the model, indicating that the selection and outcome equations are correlated, and Table 8. Parameters estimates of the mobile money and multidimensional energy poverty equations. Variables (1) (2) (3) (1) (2) (3) Burkina Faso Togo Selection equation Multidimensional energy poverty Selection equation Multidimensional energy poverty Mobile money 1/0 Mobile money not adopted Mobile money adopted Mobile money 1/0 Mobile money not adopted Mobile money adopted Age 0.0410 −0.0015 −0.0029 0.0515 −0.0005 −0.0041 (0.0072) (0.0007) (0.0022) (0.0073) (0.0011) (0.0028) Age squared −0.0004 0.0000 0.0000 −0.0006 0.0000 0.0000 (0.0001) (0.0000) (0.0000) (0.0001) (0.0000) (0.0000) Female −0.2097 0.0001 −0.0510 −0.3210 −0.0028 0.0076 (0.0472) (0.0047) (0.0136) (0.0406) (0.0076) (0.0154) Married 0.0830 −0.0178 0.0155 −0.0171 −0.0101 −0.0030 (0.0544) (0.0055) (0.0146) (0.0468) (0.0078) (0.0134) Literacy 0.5475 −0.0366 −0.0486 −0.0922 −0.0157 0.0480 (0.0543) (0.0062) (0.0177) (0.0593) (0.0099) (0.0180) Household size 0.0218 −0.0004 0.0075 −0.0254 −0.00710.0155 (0.0063) (0.0007) (0.0018) (0.0214) (0.0036) (0.0061) Household size squared −0.0002 0.0000 −0.0001 0.0014 0.0004−0.0009 (0.0001) (0.0000) (0.0000) (0.0015) (0.0003) (0.0004) Urban 0.6458 −0.1873 −0.2984 0.5819 −0.1858 −0.1795 (0.0607) (0.0077) (0.0178) (0.0445) (0.0093) (0.0240) Social capital 0.1365 0.0003 0.0188 0.2878 −0.0316 −0.0194 (0.0585) (0.0056) (0.0159) (0.0411) (0.0075) (0.0152) Self-employed 0.0241 0.0136 0.0629 0.0283 −0.0095 0.0100 (0.0514) (0.0051) (0.0138) (0.0575) (0.0104) (0.0162) Employed 0.7841 −0.1979 −0.05110.5518 −0.0328 −0.0478 (0.2220) (0.0531) (0.0293) (0.0861) (0.0220) (0.0247) Time to reach the nearest mobile money agent 5 −0.0119 −0.0133 (0.0006) (0.0018) Constant −1.3367 0.8879 −0.0069 −1.7146 0.8421 0.6706 (0.1528) (0.0140) (0.0020) (0.1502) (0.0226) (0.1028) lns0 −2.0277 −1.5748 (0.0128) (0.0212) lns1 −1.4537 −1.5545 (0.0327) (0.0386) r0 0.2942 0.4044 (0.0531) (0.1201) r1 0.6535 0.2357 (0.1167) (0.2089) sigma0 0.3164 0.2071 (0.0017) (0.0044) sigma1 0.2337 0.2113 (0.0076) (0.0081) rho0 0.2860 0.3837 (0.0488) (0.1024) rho1 0.5740 0.2315 (0.0782) (0.1977) Observations 5,013 5,013 5,013 5,045 5,045 5,045 Log likelihood 373.21424 −1699.814 Wald chi2 (11) 729.72 517.09 Prob >chi2 0.0000 0.0000 LR test of indep. eqns. : chi2(2) ¼9.83 Prob >chi2 ¼0.0000 chi2(2) ¼62.24 Prob >chi2 ¼0.0074 Robust standard errors in parentheses p<0.01, p<0.05, p<0.1. Source: Authors’computation. 16 E. D. COMPAORE ET AL. that it is appropriate to estimate these three equations jointly. Our instrumental variable is significantly and negatively correlated with mobile money adoption in both countries. Thus, as expected, an increase in distance to the nearest mobile money branch is associated with a decrease in mobile money adoption. An additional analysis of the correlation coefficients shows that rho0 is significant in the estimates carried out in both countries, which indicates that the multidimensional energy deprivation scores of non-adopters of mobile money are significantly different from those of a random household in the sample. The positive value of rho0 in fact means that the multidimensional energy deprivation scores of non-adopters of mobile money are lower than those of a random household in the sample. The rho1 is positive in the estimates for both countries but only significant in the case of Burkina Faso. This indicates that the multidimensional energy deprivation scores of Burkinabe households adopting mobile money are significantly different from those of a random household in the sample. The positive value of rho1 means that the multidimensional energy deprivation scores of households adopting mobile money are lower than those of a random household in the sample. Table 9 presents the average treatment effect estimates (ATT), which represent the overall effects of mobile money adoption on multidimensional energy deprivation scores. The results indicate that mobile money adoption significantly reduces household multidimensional energy deprivation scores by 19.32 percentage points for Burkina Faso and 20.34 percentage points for Togo. Table 9 also presents estimates of the average treatment effect on untreated individuals (ATU). The findings indicate that mobile money adoption by non-adopters would significantly reduce their multidimensional energy deprivation scores by 26.80 percentage points for the case of Burkina Faso and 15.66 percentage points for the case of Togo. These results offer an important basis for concluding that mobile money is an important alternative for reducing multidimensional energy poverty in developing countries. We also tested the robustness of the results by testing the sensitivity of the estimates using other thresholds and weightings for the MEPI. For the initial measure of energy poverty, we used the conventional 0.33 threshold. However, for the sensitivity analysis presented in Table 10 we used alternative thresholds of 0.2 and 0.5, which are consistent with the literature (Adusah-Poku & Takeuchi, 2019; Churchill & Smyth, 2020; Koomson & Danquah, 2021). The results of the alternative thresholds, presented in columns 1 and 2 for Burkina Faso and in columns 4 and 5 for Togo, are consistent with our findings. Indeed, the results indicate a positive effect of mobile money on reducing multidimensional energy poverty, which means that our results are robust to alternative thresholds in both countries. We also note that the estimates based on different thresholds (0.2, 0.33, and 0.5) yield identical results for both Burkina Faso and Togo. We also tested the sensitivity using alternative weights for energy poverty. Thus, we assign an equal weight of 0.25 to each dimension of the multidimensional energy poverty index. The results indicate a positive effect of mobile money on the reduction of multidimensional energy poverty in both countries. This indicates that the regression with the equal-weighting scheme is consistent with our main results and confirms their robustness to alternative weighting schemes. 5. Conclusion and economic policy implications This study analyses the effects of mobile money on alleviating multidimensional energy poverty. Initially, the Alkire and Foster (2011) method was employed to construct a composite Multidimensional Energy Poverty Index (MEPI). Subsequently, a linear regression model utilising instrumental variable strategy was applied to analyse the effect of mobile money adoption on energy poverty reduction. Additionally, the Table 9. Average level of multidimensional energy poverty; effects of treatment. Burkina Faso Togo Decision stage Treatment effects Decision stage Treatment effectsTo adopt Not to adopt To adopt Not to adopt 0.5757 (0.2471) 0.7689 (0.1403) ATT ¼−0:1932 (0.1209) 0 .5425 (0.0951) 0.7459 (0.1013) ATT ¼−0:2034 (0.0363) 0.5495 (0.1386) (0.8175 (0.0724) ATU ¼−0:2680 (0.0749) (0.5306 (0.1032) 0.6872 (0.1128) ATU ¼−0:1566 (0.0408) Standard deviation in parentheses p<0.01, p<0.05, p<0.1. Source: Authors’computation. COGENT ECONOMICS & FINANCE 17 Table 10. Estimates for alternative thresholds and weights for the energy poverty index. Variables Dependent variable: multidimensional energy poverty Burkina Faso Togo (1) (2) (3) (4) (5) (6) Cut-off (0.2) Cut-off (0.5) Using equal weights Cut-off (0.2) Cut-off (0.5) Using equal weights Mobile money adoption −0.2138 −0.2138 −0.2122 −0.3373 −0.3373 −0.3584 (0.0162) (0.0162) (0.0162) (0.0816) (0.0816) (0.0834) [−0.4015] [−0.4015] [−0.4067] [−0.6275] [−0.6275] [−0.6821] Age −0.0030 −0.0030 −0.0027 0.0015 0.0015 0.0019 (0.0008) (0.0008) (0.0008) (0.0015) (0.0015) (0.0015) [−0.1819] [−0.1819] [−0.1687] [0.0966] [0.0966] [0.1292] Age squared 0.0000 0.0000 0.0000 −0.0000 −0.0000 −0.0000 (0.0000) (0.0000) (0.0000) (0.0000) (0.0000) (0.0000) [0.1811] [0.1811] [0.1774] [−0.0807] [−0.0807] [−0.1142] Sex (female) −0.0163 −0.0163 −0.0147 −0.0145 −0.0145 −0.0137 (0.0055) (0.0055) (0.0056) (0.0104) (0.0104) (0.0107) [−0.0342] [−0.0342] [−0.0315] [−0.0303] [−0.0303] [−0.0293] Married −0.0072 −0.0072 −0.0118−0.0089 −0.0089 −0.0164 (0.0062) (0.0062) (0.0064) (0.0077) (0.0077) (0.0078) [−0.0146] [−0.0146] [−0.0243] [−0.0183] [−0.0183] [−0.0347] Literacy −0.0510 −0.0510 −0.0496 −0.0065 −0.0065 −0.0098 (0.0076) (0.0076) (0.0077) (0.0102) (0.0102) (0.0104) [−0.1008] [−0.1008] [−0.1001] [−0.0093] [−0.0093] [−0.0143] Household size 0.0030 0.0030 0.0021 −0.0021 −0.0021 −0.0022 (0.0006) (0.0006) (0.0007) (0.0035) (0.0035) (0.0036) [0.0722] [0.0722] [0.0507] [−0.0320] [−0.0320] [−0.0342] Household size squared −0.0000 −0.0000 −0.0000 0.0001 0.0001 0.0001 (0.0000) (0.0000) (0.0000) (0.0002) (0.0002) (0.0002) [−0.0345] [−0.0345] [−0.0196] [0.0285] [0.0285] [0.0274] Area of residence (urban) −0.2523 −0.2523 −0.2310 −0.1588 −0.1588 −0.1382 (0.0111) (0.0111) (0.0108) (0.0161) (0.0161) (0.0165) [−0.4325] [−0.4325] [−0.4040] [−0.3284] [−0.3284] [−0.2924] Social capital 0.01170.01170.0124 −0.0158 −0.0158 −0.0110 (0.0062) (0.0062) (0.0063) (0.0098) (0.0098) (0.0100) [0.0191] [0.0191] [0.0206] [−0.0332] [−0.0332] [−0.0236] Self-employed 0.0296 0.0296 0.0285 −0.0031 −0.0031 −0.0006 (0.0058) (0.0058) (0.0059) (0.0103) (0.0103) (0.0104) [0.0621] [0.0621] [0.0611] [−0.0044] [−0.0044] [−0.0009] Employed −0.1379 −0.1379 −0.1268 −0.0084 −0.0084 −0.0003 (0.0251) (0.0251) (0.0247) (0.0244) (0.0244) (0.0248) [−0.0748] [−0.0748] [−0.0702] [−0.0076] [−0.0076] [−0.0003] Constant 0.9106 0.9106 0.8749 0.8276 0.8276 0.8084 (0.0160) (0.0160) (0.0163) (0.0225) (0.0225) (0.0230) First-Stage Instrument (Time to reach the nearest mobile money agent) −0.0037 −0.0037 −0.0037 −0.0021 −0.0021 −0.0021 (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) F-statistic of first regression 612.356 612.36 612.36 108.10 108.10 108.10 Observations 5,013 5,013 5,013 5,045 5,045 5,045 R-squared 0.4857 0.4857 0.4441 0.0528 0.0528 −0.0291 Endogeneity test Durbin (score) chi2 (1) 73.2022 73.2022 70.0831 11.187 11.187 12.8642 Prob 0.0000 0.0000 0.0000 0.0008 0.0008 0.0003 Wu-Hausman F(1, 4999) /F(1,5031) 76.5231 76.5231 73.1785 11.366 11.366 13.1655 Prob 0.0000 0.0000 0.0000 0.0008 0.0008 0.0003 Test of weakness of instruments R-sq 0.3422 0.3422 0.3422 0.1159 0.1159 0.1159 Adjusted R-sq. 0.3406 0.3406 0.3406 0.1138 0.1138 0.1138 Partial R-sq 0.1504 0.1504 0.1504 0.0092 0.0092 0.0092 F(1, 5000) / F(1,5032) 612.356 612.356 612.356 108.104 108.104 108.104 Prob >F 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Weak-instrument-robust inference Anderson-Rubin Wald test F (1,5000) / F (1,5032) 189.33 189.33 184.56 17.81 17.81 19.47 p-val 0.000 0.000 0.000 0.000 0.000 0.000 (continued) 18 E. D. COMPAORE ET AL. endogenous switching regression (ESR) model was used to check the robustness of the results. The results indicate a significantly higher incidence of multidimensional energy poverty in Togo than in Burkina Faso, and a higher intensity of multidimensional energy poverty in Burkina Faso than in Togo. Moreover, the results show a positive effect of mobile money adoption on the reduction of multidimensional energy poverty. This result is robust compared to quasi-experimental methods and other sensitivity checks, such as alternative thresholds for multidimensional energy poverty. Overall, the findings indicate that mobile money contributes to the reduction in multidimensional energy poverty in Burkina Faso and Togo; however, access to formal financial services remains a challenge in developing countries. In these countries, where a large proportion of the population is excluded from using formal financial services, our results support the view that mobile money is an effective policy alternative for reducing energy poverty, thus contributing to the attainment of Sustainable Development Goal 7, which aims to ensure household access to modern energy services. Therefore, we advocate for the promotion of mobile money usage in developing countries to achieve reductions in energy poverty. Policies aimed at reducing or even exempting mobile money transactions from taxation could be an effective way of encouraging the use of mobile money in developing countries. Governments are encouraged to put in place the necessary structures to improve the coverage and quality of mobile phone networks on the one hand, and to decrease the average distance to travel to a mobile money branch on the other, especially in rural areas. Such efforts can incentivise the adoption of mobile money and mitigate the risk of energy poverty both presently and in the future. In line with Seng (2021), we will suggest that our policy-makers consider integrating financial education into the formal and nonformal education system with a particular focus on personal finance issues, such as budgeting, money management and financial planning in the context of mobile money. It is very important that policy-makers support the creation of dynamic mobile agency services that integrate the world’s poor into financial markets and enable them to use and manage their own money (Donovan, 2012). One of the limitations of this study is that our data do not allow for dynamic analysis as we are using cross-sectional data. Panel data are not available. In addition, although we assume that an increase in household income is the main channel for the impact of mobile money on reducing multidimensional energy poverty, our data do not allow us to test this hypothesis. Addressing this limitation and utilising more appropriate data, such as panel data, will serve as the basis for future research. Authors’contributions All the authors contributed to the design and development of the study. The methodological part and analysis of the results were written by COMPAORE Dimaviya Eugène and Maiga Boukar e. The introduction and literature review were written by Asmo Guira. All authors have read and approved the final manuscript. All authors accept responsibility for all aspects of their work. Disclosure statement No potential conflict of interest was reported by the author(s). Table 10. Continued. Variables Dependent variable: multidimensional energy poverty Burkina Faso Togo (1) (2) (3) (4) (5) (6) Cut-off (0.2) Cut-off (0.5) Using equal weights Cut-off (0.2) Cut-off (0.5) Using equal weights Anderson-Rubin Wald test Chi-sq (1) / Chi-sq (1) 189.83 189.83 185.04 17.85 17.85 19.52 P-val 0.000 0.000 0.000 0.000 0.000 0.000 Stock-Wright LM S statistic Chi-sq (1) 180.20 180.20 0.000 17.38 17.38 18.89 p-val 0.000 0.000 174.05 0.000 0.000 0.000 Robust standard errors in parentheses p<0.01, p<0.05, p<0.1. Standardised coefficients are in square brackets. Source: Authors’computation. COGENT ECONOMICS & FINANCE 19 Notes 1. International Energy Agency. 2. National Institute of Statistics and Demography (INSD)/Burkina Faso. 3. National Institute of Statistics and Economic and Demographic Studies (INSEED)/Togo. 4. Central Bank of West African States (BCEAO). 5. Instrumental variable (IV). About the authors Eugène Dimaviya Compaore holds a PhD in applied economics from Thomas Sankara University in Burkina Faso. Compaore’s areas of interest are gender economics, multidimensional poverty, energy poverty, financial inclusion, and women’s empowerment. Compaore is an active member of the research team on Agricultural Development and Transformation (DATA) at the Center for Economic and Social Studies, Documentation and Research (CEDRES) at Thomas Sankara University. Asmo Guira is a doctoral student in applied economics at Thomas Sankara University in Burkina Faso. Her fields of interest are rural development, agricultural policies, energy poverty, and financial inclusion. She is an active member of the Environmental Economics, Natural Resources, and Development (ERD) research team at the Center for Economic and Social Studies, Documentation and Research (CEDRES) at Thomas Sankara University. Boukar e Maiga holds a doctorate in applied economics from Thomas Sankara University in Burkina Faso. Maiga’s fields of interest are natural resources, economic growth, financial development, and financial inclusion. Boukar eis an active member of the Agricultural Development and Transformation (DATA) research team at the Center for Economic and Social Studies, Documentation and Research (CEDRES) at Thomas Sankara University. Data availability The data that support the findings of this study are available from the corresponding author, [COMPAORE Dimaviya Eugène], upon reasonable request. References Adusah-Poku, F., & Takeuchi, K. (2019). Energy poverty in Ghana: Any progress so far? Renewable and Sustainable Energy Reviews,112, 853–864. https://doi.org/10.1016/j.rser.2019.06.038 Afawubo, K., Couchoro, M. K., Agbaglah, M., & Gbandi, T. (2020). Mobile money adoption and households’vulnerability to shocks: evidence from Togo. Applied Economics,52(10), 1141–1162. https://doi.org/10.1080/00036846.2019. 1659496 Ahmad, A. H., Green, C., & Jiang, F. (2020). Mobile money, financial inclusion and development: A review with reference to African experience. Journal of Economic Surveys,34(4), 753–792. https://doi.org/10.1111/joes.12372 Ahmed, M. H., & Mesfin, H. M. (2017). The impact of agricultural cooperatives membership on the wellbeing of smallholder farmers: Empirical evidence from eastern Ethiopia. Agricultural and Food Economics,5(1), 1–20. https:// doi.org/10.1186/s40100-017-0075-z Aker, J. C., Boumnijel, R., McClelland, A., & Tierney, N. (2016). Payment mechanisms and anti-poverty programs: Evidence from a mobile money cash transfer experiment in Niger. Economic Development and Cultural Change, 65(1), 1–37. https://doi.org/10.1086/687578 Alkire, S., & Foster, J. (2011). Counting and multidimensional poverty measurement. Journal of Public Economics, 95(7–8), 476–487. https://doi.org/10.1016/j.jpubeco.2010.11.006  Alvarez-Gamboa, J., Cabrera-Barona, P., & J acome-Estrella, H. (2021). Financial inclusion and multidimensional poverty in Ecuador: A spatial approach. World Development Perspectives,22, 100311. https://doi.org/10.1016/j.wdp.2021. 100311 Apeti, A. E. (2022). Mobile money et volatilit e de la consommation des m enages. Aron, J. (2018). Mobile money and the economy: A review of the evidence. The World Bank Research Observer,33(2), 135–188. https://doi.org/10.1093/wbro/lky001 Asfaw, S., Shiferaw, B., Simtowe, F., & Lipper, L. (2012). Impact of modern agricultural technologies on smallholder welfare: Evidence from Tanzania and Ethiopia. Food Policy,37(3), 283–295. https://doi.org/10.1016/j.foodpol.2012. 02.013 Bachas, P., Gertler, P., Higgins, S., & Seira, E. (2018). Digital finance services go a long way: Transaction costs and financial inclusion. AEA Papers and Proceedings,108, 444–448. https://doi.org/10.1257/pandp.20181013 Barnes, D. F., Khandker, S. R., & Samad, H. A. (2011). Energy poverty in rural Bangladesh. Energy Policy,39(2), 894– 904. https://doi.org/10.1016/j.enpol.2010.11.014 20 E. D. COMPAORE ET AL. BCEAO. (2021). Rapport sur la situation de l’inclusion financière dans l’UEMOA au titre de l‘ann ee 2020. Bezu, S., Kassie, G. T., Shiferaw, B., & Ricker-Gilbert, J. (2014). Impact of improved maize adoption on welfare of farm households in Malawi: A panel data analysis. World Development,59, 120–131. https://doi.org/10.1016/j.worlddev. 2014.01.023 Biermann, P. (2016). How fuel poverty affects subjective well-being: Panel evidence from Germany. Oldenburg Discussion Papers in Economics No. V-395. 16. Birol, F. (2007). Energy economics: a place for energy poverty in the agenda? The Energy Journal,28(3), iv–i6. https:// doi.org/10.5547/ISSN0195-6574-EJ-Vol28-No3-1 Boardman, B. (1991). Fuel poverty: From cold homes to affordable warmth. Belhaven Press. Bouzarovski, S., & Petrova, S. (2015). A global perspective on domestic energy deprivation: Overcoming the energy poverty–fuel poverty binary. Energy Research & Social Science,10,31–40. https://doi.org/10.1016/j.erss.2015.06.007 Bridge, B. A., Adhikari, D., & Fontenla, M. (2016). Electricity, income, and quality of life. The Social Science Journal, 53(1), 33–39. https://doi.org/10.1016/j.soscij.2014.12.009 Bukari, C., & Koomson, I. (2020). Adoption of mobile money for healthcare utilization and spending in rural Ghana (pp. 37–60). Springer Singapore. Cameron, A. C., & Trivedi, P. K. (2010). Microeconometrics using stata (Vol. 2). Stata press. Casillas, C. E., & Kammen, D. M. (2010). The energy-poverty-climate nexus. Science (New York, N.Y.),330(6008), 1181– 1182. https://doi.org/10.1126/science.1197412 Churchill, S. A., & Marisetty, V. B. (2019). Financial inclusion and poverty: A tale of forty-five thousand households. Applied Economics,52(16), 1777–1788. https://doi.org/10.1080/00036846.2019.1678732 Churchill, S. A., Nuhu, S. A., & Smyth, R. (2020). Financial inclusion and poverty: Micro-level evidence from Nigeria. In Moving from the millennium to the sustainable development goals (pp. 11–36). Springer. Churchill, S. A., & Smyth, R. (2020). Ethnic diversity, energy poverty and the mediating role of trust: Evidence from household panel data for Australia. Energy Economics,86, 104663. https://doi.org/10.1016/j.eneco.2020.104663 Coulibaly, Y. (2023). Can mobile money adoption affect environmental sustainability in energy transition context?. Crentsil, A. O., Asuman, D., & Fenny, A. P. (2019). Assessing the determinants and drivers of multidimensional energy poverty in Ghana. Energy Policy,133, 110884. https://doi.org/10.1016/j.enpol.2019.110884 Dabla-Norris, E., Ji, Y., Townsend, R. M., & Unsal, D. F. (2021). Distinguishing constraints on financial inclusion and their impact on GDP, TFP, and the distribution of income. Journal of Monetary Economics,117,1–18. https://doi. org/10.1016/j.jmoneco.2020.01.003 Day, R., Walker, G., & Simcock, N. (2016). Conceptualising energy use and energy poverty using a capabilities framework. Energy Policy,93, 255–264. https://doi.org/10.1016/j.enpol.2016.03.019 Demirg€ uc¸-Kunt, A., Klapper, L., & Singer, D. (2017). Financial inclusion and inclusive growth: A review of recent empirical evidence. The World Bank. Demirg€ uc¸-Kunt, A., Klapper, L., & Singer, D. (2018). Household finance and economic development. In Handbook of finance and development (pp. 534–549). Edward Elgar Publishing. Di Falco, S., Veronesi, M., & Yesuf, M. (2011). Does adaptation to climate change provide food security? A micro-perspective from Ethiopia. American Journal of Agricultural Economics,93(3), 829–846. https://doi.org/10.1093/ajae/ aar006 Djahini-Afawoubo, D. M., Couchoro, M. K., & Atchi, F. K. (2023). Does mobile money contribute to reducing multidimensional poverty? Technological Forecasting and Social Change,187, 122194. https://doi.org/10.1016/j.techfore. 2022.122194 Dogan, E., Madaleno, M., & Taskin, D. (2021). Which households are more energy vulnerable? Energy poverty and financial inclusion in Turkey. Energy Economics,99, 105306. https://doi.org/10.1016/j.eneco.2021.105306 Donovan, K. (2012). Mobile money for financial inclusion. Information and Communications for Development,61(1), 61–73. Duflo, E. (2020). Field experiments and the practice of policy. American Economic Review,110(7), 1952–1973. https:// doi.org/10.1257/aer.110.7.1952 Gonzalez, C. G. (2016). Energy poverty and the environment. In International energy and poverty: The emerging contours (pp. 113–130). Haushofer, J., & Shapiro, J. (2016). The short-term impact of unconditional cash transfers to the poor: Experimental evidence from Kenya. The Quarterly Journal of Economics,131(4), 1973–2042. https://doi.org/10.1093/qje/qjw025 Hills, J. (2012). Getting the measure of fuel poverty: Final Report of the Fuel Poverty Review. Hjelm, L., Handa, S., de Hoop, J., & Palermo, T. (2017). Poverty and perceived stress: Evidence from two unconditional cash transfer programs in Zambia. Social Science & Medicine (1982),177, 110–117. https://doi.org/10.1016/j.socscimed.2017.01.023 Hosan, S., Sen, K. K., Rahman, M. M., Karmaker, S. C., Chapman, A. J., & Saha, B. B. (2023). Evaluating the mediating role of energy subsidies on social well-being and energy poverty alleviation in Bangladesh. Energy Research & Social Science,100, 103088. https://doi.org/10.1016/j.erss.2023.103088 INSD. (2021). Enqu^ ete Harmonis ee sur les Conditions de Vie des M enages (EHCVM) de 2018. Rapport g en eral du. Burkina Faso. INSEED. (2020). Profil de pauvrete. Rapport final du Togo. COGENT ECONOMICS & FINANCE 21 International Energy Agency (IEA), International Renewable Energy Agency, United Nations Statistics Division, World Bank, & World Health Organisation. (2020). Tracking SDG 7 : The Energy Progress Report (2020). Islam, A. H., Basher, S. A., & Enamul Haque, K. (2022). The impact of mobile money on long-term poverty: evidence from Bangladesh. Journal of Social and Economic Development,24(2), 436–455. https://doi.org/10.1007/s40847-02200194-0 Jack, W., & Suri, T. (2014). Risk sharing and transactions costs: Evidence from Kenya’s mobile money revolution. American Economic Review,104(1), 183–223. https://doi.org/10.1257/aer.104.1.183 Jensen, R. (2007). The digital provide: Information (technology), market performance, and welfare in the south Indian fisheries sector. The Quarterly Journal of Economics,122(3), 879–924. https://doi.org/10.1162/qjec.122.3.879 Kahouli, S. (2020). An economic approach to the study of the relationship between housing hazards and health: the case of residential fuel poverty in France. Energy Economics,85, 104592. https://doi.org/10.1016/j.eneco.2019. 104592 Khandker, S. R., Koolwal, G. B., & Samad, H. A. (2010). Handbook on impact evaluation: Quantitative methods and practices. World Bank Publications. Khanna, R. A., Li, Y., Mhaisalkar, S., Kumar, M., & Liang, L. J. (2019). Comprehensive energy poverty index: Measuring energy poverty and identifying micro-level solutions in south and Southeast Asia. Energy Policy,132, 379–391. https://doi.org/10.1016/j.enpol.2019.05.034 Koomson, I., & Danquah, M. (2021). Financial inclusion and energy poverty: Empirical evidence from Ghana. Energy Economics,94, 105085. https://doi.org/10.1016/j.eneco.2020.105085 Koomson, I., Villano, R. A., & Hadley, D. (2020). Effect of financial inclusion on poverty and vulnerability to poverty: Evidence using a multidimensional measure of financial inclusion. Social Indicators Research,149(2), 613–639. https://doi.org/10.1007/s11205-019-02263-0 Kushlev, K., & Proulx, J. (2016). The social costs of ubiquitous information: Consuming information on mobile phones is associated with lower trust. PloS One,11(9), e0162130. https://doi.org/10.1371/journal.pone.0162130 Ky, S., Rugemintwari, C., & Sauviat, A. (2018). Does mobile money affect saving behaviour? Evidence from a developing country. Journal of African Economies,27(3), 285–320. https://doi.org/10.1093/jafeco/ejx028 Labonne, J., & Chase, R. S. (2009). The power of information: The impact of mobile phones on farmers’welfare in the Philippines. Available at. In World Bank policy research working paper no. 4996.https://ssrn.com/abstract=1435202 Leach, G. (1992). The energy transition. Energy Policy,20(2), 116–123. https://doi.org/10.1016/0301-4215(92)90105-B Lee, J. N., Morduch, J., Ravindran, S., Shonchoy, A., & Zaman, H. (2021). Poverty and migration in the digital age: Experimental evidence on mobile banking in Bangladesh. American Economic Journal: Applied Economics,13(1), 38–71. https://doi.org/10.1257/app.20190067 Lewis, P. (1982). Fuel poverty can be stopped. National Right to Fuel Campaign. Llorca, M., Rodriguez-Alvarez, A., & Jamasb, T. (2020). Objective vs. subjective fuel poverty and self-assessed health. Energy Economics,87, 104736. https://doi.org/10.1016/j.eneco.2020.104736 Lokshin, M., & Sajaia, Z. (2004). Maximum likelihood estimation of endogenous switching regression models. The Stata Journal: Promoting Communications on Statistics and Stata,4(3), 282–289. https://doi.org/10.1177/ 1536867X0400400306 Matekenya, W., Moyo, C., & Jeke, L. (2020). Financial inclusion and human development: Evidence from Sub-Saharan Africa. Development Southern Africa,38(5), 683–700. https://doi.org/10.1080/0376835X.2020.1799760 Mdluli, P., & Dunga, S. (2021). Determinants of poverty in South Africa using the 2018 general household survey data. Journal of Poverty,26(3), 197–213. Munyegera, G. K., & Matsumoto, T. (2016). Mobile money, remittances, and household welfare: Panel evidence from Rural Uganda. World Development,79, 127–137. https://doi.org/10.1016/j.worlddev.2015.11.006 Naito, H., Ismailov, A., & Kimaro, A. B. (2021). The effect of mobile money on borrowing and saving: Evidence from Tanzania. World Development Perspectives,23, 100342. https://doi.org/10.1016/j.wdp.2021.100342 Nan, W., Zhu, X., & Lynne Markus, M. (2021). What we know and don’t know about the socioeconomic impacts of mobile money in Sub-Saharan Africa: A systematic literature review. The Electronic Journal of Information Systems in Developing Countries,87(2), e12155. https://doi.org/10.1002/isd2.12155 Nanziri, E. L. (2016). Financial inclusion and welfare in South Africa: Is there a gender gap? Journal of African Development,18(2), 109–134. https://doi.org/10.5325/jafrideve.18.2.0109 Njiru, C. W., & Letema, S. C. (2018). Energy poverty and its implication on standard of living in Kirinyaga, Kenya. Journal of Energy,2018,1–12. https://doi.org/10.1155/2018/3196567 Nonvide, G. M. A. (2023). Impact of information and communication technologies on agricultural households’welfare in Benin. Telecommunications Policy,47(6), 102570. Nussbaumer, P., Bazilian, M., & Modi, V. (2012). Measuring energy poverty: Focusing on what matters. Renewable and Sustainable Energy Reviews,16(1), 231–243. https://doi.org/10.1016/j.rser.2011.07.150 Oezcan, K. M., Guelay, E., & Uecdogruk, S. (2013). Economic and demographic determinants of household energy use in Turkey. Energy Policy,60, 550–557. https://doi.org/10.1016/j.enpol.2013.05.046 Ogwumike, F. O., & Ozughalu, U. M. (2016). Analysis of energy poverty and its implications for sustainable development in Nigeria. Environment and Development Economics,21(3), 273–290. https://doi.org/10.1017/ S1355770X15000236 22 E. D. COMPAORE ET AL. Omar, M. A., & Inaba, K. (2020). Does financial inclusion reduce poverty and income inequal ity in developing countries? A panel data analysis. Journal of Economic Structures,9,1–25. Oum, S. (2019). Energy poverty in the Lao PDR and its impacts on education and health. Energy Policy,132, 247–253. https://doi.org/10.1016/j.enpol.2019.05.030 Park, C. Y., & Mercado, R. V. Jr, (2018). Financial inclusion, poverty, and income inequality in developing Asia. The Singapore Economic Review,63(01), 185–206. https://doi.org/10.1142/S0217590818410059 Pereira, M. G., Freitas, M. A. V., & da Silva, N. F. (2011). The challenge of energy poverty: Brazilian case study. Energy Policy,39(1), 167–175. https://doi.org/10.1016/j.enpol.2010.09.025 Phoumin, H., & Kimura, F. (2019). The impacts of energy insecurity on household welfare in Cambodia: Empirical evidence and policy implications. Economic Modelling,82,35–41. https://doi.org/10.1016/j.econmod.2019.09.024 Polloni-Silva, E., da Costa, N., Moralles, H. F., & Sacomano Neto, M. (2021). Does financial inclusion diminish poverty and Inequality? A panel data analysis for Latin American countries. Social Indicators Research,158(3), 889–925. https://doi.org/10.1007/s11205-021-02730-7 Practical Action. (2014). Poor People’s Energy Outlook 2014: Key messages on energy for poverty alleviation. Practical Action Publishing. Pye, S., Dobbins, A., Baffert, C., Brajkovi c, J., Deane, P., & De Miglio, R. (2015). Addressing energy poverty and vulnerable consumers in the energy sector across the EU. L’Europe En Formation,n378(4), 64–89. https://doi.org/10. 3917/eufor.378.0064 Quandt, A., Salerno, J. D., Neff, J. C., Baird, T. D., Herrick, J. E., McCabe, J. T., Xu, E., & Hartter, J. (2020). Mobile phone use is associated with higher smallholder agricultural productivity in Tanzania, East Africa. PloS One,15(8), e0237337. https://doi.org/10.1371/journal.pone.0237337 Reddy, S. B., Balachandra, P., & Hippu, S. K. (2010). Universalization of access to modern energy services in Indian households. Energy Policy Analysis,37(11), 4645–4657. https://doi.org/10.1016/j.enpol.2009.06.021 Reilly, J. (2015). Energie et d eveloppement dans les pays  emergents. Revue D’ economie du D eveloppement,23(3), 19– 41. https://doi.org/10.3917/edd.293.0019 Riley, E. (2018). Mobile money and risk sharing against village shocks. Journal of Development Economics,135,43–58. https://doi.org/10.1016/j.jdeveco.2018.06.015 Rosenthal, J., Quinn, A., Grieshop, A. P., Pillarisetti, A., & Glass, R. I. (2018). Clean cooking and the SDGs: Integrated analytical approaches to guide energy interventions for health and environment goals. Energy for Sustainable Development: The Journal of the International Energy Initiative,42, 152–159. https://doi.org/10.1016/j.esd.2017.11. 003 Sadath, A. C., & Acharya, R. H. (2017). Assessing the extent and intensity of energy poverty using multidimensional energy poverty index: empirical evidence from households in India. Energy Policy,102, 540–550. https://doi.org/10. 1016/j.enpol.2016.12.056 Sambodo, M. T., & Novandra, R. (2019). The state of energy poverty in Indonesia and its impact on welfare. Energy Policy,132, 113–121. https://doi.org/10.1016/j.enpol.2019.05.029 Sani Ibrahim, S., Ozdeser, H., & Cavusoglu, B. (2019). Financial inclusion as a pathway to wel fare enhancement and income equality: Micro-level evidence from Nigeria. Development Southern Africa,36(3), 390–407. https://doi.org/ 10.1080/0376835X.2018.1498766 Sarma, M., & Pais, J. (2011). Financial inclusion and development. Journal of International Development,23(5), 613– 628. https://doi.org/10.1002/jid.1698 Sen, K. K., Karmaker, S. C., Hosan, S., Chapman, A. J., Uddin, M. K., & Saha, B. B. (2023). Energy poverty alleviation through financial inclusion: Role of gender in Bangladesh. Energy,282, 128452. https://doi.org/10.1016/j.energy. 2023.128452 Seng, K. (2021). The mobile money’s poverty-reducing promise: evidence from Cambodia. World Development Perspectives,22, 100310. https://doi.org/10.1016/j.wdp.2021.100310 Sharma, D. (2016). Nexus between financial inclusion and economic growth: evidence from the emerging Indian. Journal of Financial Economic Policy,8(1), 13–36. https://doi.org/10.1108/JFEP-01-2015-0004 Staiger, D., & Stock, J. (1997). Instrumental variables regression with weak instruments. Econometrica,65(3), 557–586. Stein, L. C., & Yannelis, C. (2019). Financial inclusion, human capital, and wealth accumula tion: Evidence from the freedman’s savings bank. The Review of Financial Studies,33(11), 5333–5377. https://doi.org/10.1093/rfs/hhaa013 Suri, T., & Jack, W. (2016). The long-run poverty and gender impacts of mobile money. Science (New York, N.Y.), 354(6317), 1288–1292. https://doi.org/10.1126/science.aah5309 Verkaart, S., Munyua, B. G., Mausch, K., & Michler, J. D. (2017). Welfare impacts of improved chickpea adoption: A pathway for rural development in Ethiopia? Food Policy,66,50–61. https://doi.org/10.1016/j.foodpol.2016.11.007 Wachtel, P. (2001). Growth and finance: What do we know and how do we know it? International Finance,4(3), 335– 362. https://doi.org/10.1111/1468-2362.00077 Wang, X., Yang, W., Ren, X., & Lu, Z. (2023). Can financial inclusion affect energy poverty in China? Evidence from a spatial econometric analysis. International Review of Economics & Finance,85, 255–269. https://doi.org/10.1016/j. iref.2023.01.020 Wooldridge, J. M. (2002). Econometric analysis of cross section and panel data. The MIT Press. COGENT ECONOMICS & FINANCE 23