Policy Research Working Paper 9699 Economics of Distributed Photovoltaics An Illustration from Bangladesh Govinda R Timilsina Development Economics Development Research Group June 2021 Policy Research Working Paper 9699 Abstract Distributed photovoltaics are a growing technology for grid consumers’ perspective, the study finds that the econom- electricity consumers in low- and middle-income coun- ics of distributed photovoltaics depends on the difference tries due to declining costs and government support. In in electricity production costs between the distributed Bangladesh, distributed photovoltaics iare part of broader photovoltaics and the electricity utility, transmission and solar and consumer programs. This study analyzes the eco- distribution loss, and feed-in arrangements. The study also nomics of stylized grid-connected residential, commercial, reveals that a distributed photovoltaics do not necessarily and industrial distributed photovoltaics in Bangladesh, cause loss to the national electricity utility if they replaces considering a year of hourly patterns of solar irradiation expensive oil-fired generation. From a national or societal and electricity exchanges between the distributed photo- perspective, distributed photovoltaics are beneficial even voltaics owners and the electricity utilities. The economics if their positive environmental effects are not taken into vary between different stakeholders—distributed photo- account. The environmental benefits further improve the voltaics owners, electricity utilities, and society. From the economics of distributed photovoltaics. This paper is a product of the Development Research Group, Development Economics. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The author may be contacted at gtimilsina@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Economics of Distributed Photovoltaics: An Illustration from Bangladesh1 Govinda R Timilsina2 Key Words: Renewable energy, Solar power, Photovoltaics, Distributed PV, Bangladesh, Economics of solar PV JEL Classification: Q42 1 The author would like to thank Alan Lee, Tu Chi Nguyen and Zuzana Dobrotkova for their valuable comments and suggestions. Thomas Flochel, Sandra Laura Chavez Velazquez and Amit Jain also provided feedback during the development of the methodology for the study. Rachel Linda Fox, Bipul Singh and Debabrata Chattopadhyay provided important data for the study. The paper was presented in the 2021 annual international conference (virtual) of the International Association for Energy Economics. World Bank’s Energy Sector Management Assistant Program (ESMAP) partially funded the study. The views and interpretations are of authors and should not be attributed to the World Bank Group and the organizations they are affiliated with. 2 Senior Research Economist, Sustainable Development and Infrastructure Unit, Development Research Group (gtimilsina@worldbank.org) Economics of Distributed Photovoltaics: An Illustration from Bangladesh 1. Introduction The deployment of solar energy technologies, such as solar photovoltaics (PV), is rapidly increasing in many countries worldwide. The sharp drops in its costs through technological improvements and economy of scale in its deployment and continuous policy support contributed to the growth of solar technologies. Historically, most PV installed globally has been utility-scale ground-mounted systems for transmission to consumers far away. In contrast, small-scale PV technologies, such as solar home systems (SHS) 3 retain popularity in areas where access to grid electricity is absent or unreliable. However, PV systems can also be connected to a distribution grid or a consumer connected to a grid. Systems of up to several megawatts (MW) in size can be installed on rooftops, as canopies above irrigation canals, or in floating arrays on industrial ponds. Such distributed PV (DPV) systems are irrevocably changing power systems around the world especially as power may flow in two directions between consumers and the grid (ESMAP, 2021). DPV owners/operators can reduce their net electricity bill through ‘net-metering’ or ‘net-billing’ (EnergySage 2021; Qadrdan et al. 2018). The attraction of the private sector (households, commercial and industrial entrepreneurs) and attention of the public sector towards DPV will depend on the economics of DPV. If the private sector stakeholders anticipate net gains from DPVs, they will install it. Governments or government-owned electricity utilities find it beneficial if the overall societal gains (i.e., gains including those through reductions of CO2 and local air pollutants) are higher than the costs of DPV, including the lost sale of utility-supplied electricity. Besides, DPV is expected to deliver other benefits such as backup supply, especially when the electricity grid is unreliable, reducing non-technical losses (i.e., electricity theft) in areas where such malpractices are commonplace (ESMAP, 2021). However, the economics of DPV would differ across jurisdictions depending upon several economic and technical characteristics of DPV and the electricity grids. The technical characteristics include solar energy profiles, electricity load or demand patterns, electricity generation mix of the grid, transmission and distribution losses of the grid, and system losses of the DPV. Economic characteristics include grid electricity tariffs, costs of electricity produced from DPV and valuation of climate change, and other environmental benefits. Economics of DPV 3 Single or multiple solar panels installed in residential or commercial buildings. 2 is also sensitive to regulatory guidelines, for example, whether the grid accepts electricity supplied by DPV for all hours or only selected hours; what electricity tariffs the grid offers to the DPV owners. Therefore, a series of studies analyzing the economics of DPV from different stakeholders' perspectives are necessary for different jurisdictions. While a few studies discussed in the following paragraphs are available in the literature, this study aims to enrich the literature through an illustrative analysis considering the case of a developing country, Bangladesh. Vilaça Gomes et al. (2018), Holdermann et al. (2014) and Mitscher and Rüther (2012) present techno-economic analysis of DPV in Brazil, where the net metering policy has been in place since 2012 to promote DPV. Through the techno-economic analysis, Vilaça Gomes et al. (2018) find that DPV, of the capacity of about 3.7 GW, would be economically viable for 4 million residential buildings in the next 25 years. The study, however, has several limitations. It covers only the residential sector. The analysis offers investors’ perspectives only. It does not analyze the impacts of the DPV on the distribution utilities and the country as a whole (no climate change and environmental benefits are accounted for). The profile of solar energy and load curves are aggregated or averaged for a year, whereas both solar profiles and load vary significantly across months or seasons in a year. Holdermann et al. (2014) assess the economic viability of DPV in the residential and commercial sectors to check if the net metering regulation introduced attracts DPV investors. Their results show DPV is not economically viable unless the discount rate is 6% or lower and capital costs of DPV are 20% lower than what they use. However, the study has similar limitations as in Vilaça Gomes et al. (2018) except for the sectoral coverage. Ahmad (2021) analyzes the effects of DPV on Lebanon’s electricity utility, which is highly underperforming for various reasons. It shows that DPV helps reduce the financial losses that the national utility is currently experiencing because electricity produced from DPV is much cheaper than the current electricity grid. The study also finds that the national utility may not benefit from the DPV in the future if the grid performance improves and fuel price remains at a low level. The analysis is highly simplified. It neither considers hourly profiles of solar nor the system load. All calculations are done at an aggregated level. Moreover, the study assumes that the DPV replaces the long-run marginal costs of the grid; in reality, DPV replaces short-run marginal costs because it substitutes only energy (MWh) from the grid; being an intermittent resource, it cannot substitute the capacity (MW) of the grid. 3 This study adds additional insights on the economics of DPV through a study for Bangladesh and contributes to the methodology for analyzing the economics of DPV on several fronts. First, it considers hourly load profiles representing each month and distinguishing load profiles of weekdays and weekends/holidays. Instead of considering a single average solar energy profile for a year, it accounts for seasonal variations of solar energy availability by considering solar profiles each month. Thus, this study captures the load and generation characteristics more precisely than existing studies. Second, this study assesses the economics of DPV from different perspectives – investor’s perspective, utility’s perspective, and national perspective, whereas existing studies consider only one perspective, either investors’ perspective (e.g., Vilaça Gomes et al. 2018; Holdermann et al. 2014) or utility’s perspective (e.g., Ahmad (2021). Third, it also distinguishes analysis across the customer category (residential, commercial, and industrial); existing studies mainly consider aggregate customers. Fourth, this study considers variations in several key variables (e.g., capital costs of DPV, size of DPV with respect to the load, discount rate, electricity tariffs offered to excess DPV electricity). This is not the case in the existing studies. The paper is organized as follows. Section 2 presents the detailed methodology, followed by data sources and assumptions in Section 3. Results and analysis are presented in Section 4. Finally, key conclusions and policy discussions are presented in Section 5. 2. Methodology 2.1 Electricity generation from DPV The electricity generation capacity of a DPV in a sector r in year y is estimated as follows: , , = ∗ , (1) , Where TCAPr,y is the total installed capacity of DPV in sector r in year y. Ar,y is the total area available for the installation of DPV in sector r. PNAr,y and PNRr,y are the area of a standard solar panel and its power rating, respectively in year y. For the commercial and industrial sectors, the total area is directly provided. In the residential sector, it is calculated by multiplying the number of households installing DPV and the average area available in each household for the DPV. The hourly solar electricity produced from the DPV in sector r in a month m of a year y (ENr,h,d,m,y) is given as: 4 ,ℎ,,, = , ∗ ℎ,,, (2) 0 ≤ fr,h,d,m,y ≤ 1 Where fr,h,d,m,y is the fraction of the peak rating of the solar panel in a given hour (h) of a day (d) in a month (m) in a year (y). When the solar panels receive the maximum amount of solar irradiation, f r,h,d,m,y is equal to 1. When there is no solar irradiation (i.e., night time) fr,h,d,m,y is equal to 0. The total electricity generation from the DPV by sector r in year y is the sum of electricity generation across the hours in a day, days in a month, and months in a year: , = ∑ℎ,, ,ℎ,,, (3) Total electricity generation from the DPV in a country in year y is the sum of electricity generation across the sectors in the given year: = ∑ , (4) 2.2 Electricity load profile As far as detailed demand-side data is available for all appliances, the load curve is estimated as follows: ,ℎ,,, = ∑ ,,ℎ,,, ∗ ,,ℎ,,, (5) Loadr,h,d,m,y is the load due to all appliances and devices in sector r in a given hour h, in a day d, in a month m, and in a year y. ALa,r,h,d,m,y and ALFa,r,h,d,m,y are, respectively, appliance name-plate load and appliances load factor in the given time slot. The appliance load factor represents the fraction of its name-plate load drawn in a given time slot. Many appliances, for example, electric motors, do not necessarily run at their full load in a given time slot. Some appliances do not draw any electricity in a certain hour, such as a light bulb that draws zero electricity during daylight. Even in the nighttime, not all light bulbs in a house are switched on. A detailed survey of appliances or energy audit is needed to estimate the load profiles DPV owners. For the electricity system load profiles, data are available at the load dispatching centers of electricity utilities or independent system operators. If the detailed survey is not possible due to 5 time and resource constraints, the load profiles of DPV owners are assumed to be identical to that of the system load profile. If daily system load curves are not available for all 365 days, it is recommended to have 24 load curves – one load curve for weekdays for each month and other load curves for weekends and holidays for each month. 2.3 Electricity exchange between the DPV and electricity utility The DPV owners are assumed to use their generation from the DPV to meet their load as long as the generation is available. If their load is higher than their own generation in a given hour, they continue to fill the gap using utility or grid supply. If their DPV generation exceeds their load, the excess electricity is sold to the utility. It is also possible that DPV owners could use the excess electricity for charging batteries and selling the stored electricity in peak hours. However, this is possible only when: (a) they have storage systems (batteries) and (b) the utility uses a two-part or real-time tariff system. Let us ignore the storage option. The amount of utility electricity sales offset by DPV (OFFEr,h,d,m,y) supplied in a given time slot in sector r in year y is: ,ℎ,,, = ,ℎ,,, ∗ (1 − ) if ENr,h,d,m,y *(1-dpvl) < Loadr,h,d,m,y (6a) ,ℎ,,, = ,ℎ,,, if ENr,h,d,m,y*(1-dpvl) > Loadr,h,d,m,y (6b) Where dpvl is the own consumption of DPV system (or DPV system loss). The electricity sale of utility to consumers in the presence of DPV (NEDCr,h,d,m,y) is: ,ℎ,,, = ,ℎ,,, − ,ℎ,,, *(1-dpvl) if ENr,h,d,m,y *(1-dpvl) < Loadr,h,d,m,y (7a) ,ℎ,,, = 0 if ENr,h,d,m,y *(1-dpvl) > Loadr,h,d,m,y (7b) The sale of DPV electricity to the utility (NECDr,h,d,m,y) is: ,ℎ,,, = ,ℎ,,, ∗ (1 − dpvl) − ,ℎ,,, if ENr,h,d,m,y*(1-dpvl) > Loadr,h,d,m,y (8a) ,ℎ,,, = 0 if ENr,h,d,m,y *(1-dpvl) < Loadr,h,d,m,y (8b) 6 2.4 Savings of transmission and distributional (T&D) loss due to DPV When the DPV owners use their generation for meeting their load fully, it saves both transmission and distribution losses of the grid. When the DPV sells electricity to the utility it saves transmission loss only because it also needs the distribution system to sell its excess electricity to the utility. Its savings from the loss reduction (SLr,h,d,m,y) would be: ,ℎ,,, ,ℎ,,, = ∗ ( + )) if ENr,h,d,m,y*(1-dpvl) < Loadr,h,d,m,y (9a) 1−− ,ℎ,,, ,ℎ,,, = ∗ if ENr,h,d,m,y *(1-dpvl) > Loadr,h,d,m,y (9b) 1− Where ‘tl’ and ‘dl’ are, respectively, transmission and distribution losses of the utility system expressed as fractions of total generation from the utility. 2.5 Economic impacts on DPV owners or investors The gains or loss for DPV owners would be the difference in electricity costs in the absence of DPV and in the presence of DPV. They can also benefit by selling the excess electricity from the DPV. Note that they have to account for the costs of electricity generation from DPV, which is given as follows: = + + (10a) with ∗∗1000 ∗1000 = ∗24∗365 (10b) = ∗24∗365 (10c) {∗(1+) } , = [{(1+)}−1] (10d) = ( (10e) , ∗24∗365) DPVCr is the unit cost of electricity generation ($/kWh) from a DPV installed in sector r; it is also referred to as Levelized cost. 4 OC is the overnight construction cost (or lump-sum investment) of the DPV expressed in terms of capacity ($/kW), and CAF is the capacity availability factor of the DPV. FXC is the annual fixed cost of the DPV expressed in terms of capacity ($/kW). CRF is the capacity recovery factor that converts the cost expressed in terms of capacity to the corresponding 4 This is a simple formula which is commonly presented in studies calculating LCOE; please see, for example, Timilsina et al. (2020). 7 cost in terms of energy. ‘dr’ is the discount rate (or cost of capital), and ‘n’ is the economic life of the DPV. The annual gain or loss to DPV owner in sector r in year y (DPVGr,y) is calculated as follows: , = {∑ℎ,, ,ℎ,,, − ,ℎ,,, } ∗ ,ℎ,,, + ∑ℎ,, ,ℎ,,, ∗ , − ∑ℎ,, ,ℎ,,, ∗ (11) UP is the utility tariff (utility electricity price paid by consumers). DPVP is the price paid by the utility for DPV electricity. Note that DPVP could be equal to UP or DPVC or something else depending upon the agreement between the DPV owners and the utility. The gain or loss to DPV owners depends on the differential between the UP and DPVC, and also DPVP. As shown in equations 10, DPVC is sensitive to the discount rate and capital costs of DPV. If DPVC is lower than UP, they are likely to gain. On the other hand, if DPVC is higher than the UP they tend to lose. In many countries, where industrial and commercial consumers are cross-subsidizing the residential consumers, DPV owners in the commercial and industrial sectors are expected to gain. On the other hand, residential DPV owners may not gain because the cost of electricity from DPV would be expensive unless subsidized compared to the subsidized electricity tariff they pay to the utility. The annual gain of DPV owners combining all sectors or consumers in a year is the sum of the gain that occurred in each sector considered: = ∑ , (12) 2.6 Impacts on the utility Whether a utility gains or loses due to the DPV depends on whether it is operating in profit or loss. If it is operating in loss, DPV could shed some of its losses, and it will be better off (reduction in existing loss). On the other hand, if it is a profit-making entity, the DPV would cause it to lose because of its markets offset by the DPVs. The loss to the utility includes its market loss as the DPV offsets the utility’s electricity markets. On the other hand, it does not need to generate the amount of electricity offset by the DPV and also the associated value of T&D losses. One critical question here is: does the DPV avoid only the costs of generation (i.e., fuel costs) of the unit operating at the margin in the given time slot (or short-run marginal cost) or the entire cost of generation facilities operating at the margin (long-run marginal cost)? We assume that if the DPV 8 does not have a storage facility, it replaces the short-run marginal costs. The gain or loss of utility (ULGy) is calculated as follows: = + − (13) If ULG is negative, the utility loses, if ULG is positive, it gains. GSLR is gain due to system (T&D) loss reduction, GFCS is gain from fuel costs savings and LULS is loss due to reduction in utility electricity sales. These are given as follows: = ∑,ℎ,, ,ℎ,,, ∗ ,ℎ,,, (14a) ,ℎ,,, ,ℎ,,, = ∑,ℎ,,{ (1−−) + (1−) } ∗ ℎ,,, (14b) = ∑,ℎ,,{,ℎ,,, + ,ℎ,,, } ∗ ,ℎ,,, (15) FC is the cost of fuel used for power generation in the given time slot (h,d,m,y). 2.7 Gain at the national level without accounting for environmental benefits The gain for a country without accounting for environmental or health benefits of DPV would be the savings of its subsidy to the utility as long as it does not provide a subsidy to the DPV. Suppose the government still provides a subsidy to the DPV. In that case, the national benefits depend on the size of subsidy it provides to the utility and DISCOM vs. the size of subsidy it provides to the DPV. The national gain or loss due to the DPV (NLG) is calculated as follows: ,ℎ,,, ,ℎ,,, = ∑,ℎ,,( + (1−) ) ∗ ℎ,,, − ∑,ℎ,, ,ℎ,,, ∗ (1−−) (16) SMC is the short-run marginal costs. If the utility receives a direct subsidy of ‘sy’ for a unit of electricity, then SMC is given as: ℎ,,, = ℎ,,, + (17) If the government provides subsidies to the utility and does not need to provide subsidies to the DPV and if the costs of DPV are lower than the utility electricity tariff, the nation gains. If the government has to provide subsidies to the DPV, yet the level of subsidies ($/kWh) is smaller than that it provides to the utility, the government still gains. However, if the size of subsidy to DPV is 9 higher than that the government provides to the utility, there would be a net loss at the national level. Obviously, the nation gains if DPV costs are lower than the cost of electricity from the grid. 2.8 Gain to society (gain at the national level including other benefits) The environmental gains (GENV) are calculated as: = ∑,ℎ,,, ℎ,,, ∗ , ∗ , (18) where COEFp are the emission coefficients of emission type p (CO2, SO2, NOx, SPM) of electricity generated by fuel type FCON, in a given time slot. It is expressed as tons per unit of electricity (GWh). ENVPp is the value of avoidance of the emissions of type p ($/tons). Emission coefficients for a given time slot could be the weighted average of all generations in that slot or that of the fuel at the margin which is likely to be replaced by the DPV. The total benefit at the national level, including environmental benefits (TNL) is given as: = + (19) 3. Data and Assumptions 3.1 DPV-related data The primary data related to DPV are daily solar profiles, technical characteristics of DPVs, and economic characteristics of DPVs to be installed. 3.1.1 Hourly solar energy profile The daily and seasonal solar profiles are the main determinants of electricity generation from a given capacity of DPV installed. Figure 1 presents the average daily solar profile by month for Bangladesh available from Global Solar Atlas (GSA) (2020). All values presented in Figure 1 are normalized; they are the fractions of maximum values of direct normal solar irradiation (721 Watt/m2) that occurs from 11 AM to 12 PM in December in a year. 10 Figure 1. Solar profiles of Bangladesh 1.00 0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 January February March April May June July August September October November December Source: GSA (2020). Note: Solar radiations vary across locations. However, Bangladesh is a small country in terms of its size (area) and all regions in the country are in the same climatic zone. Therefore, the solar profile, which refers to the Capital city Dhaka, can be interpreted as the representative (or average) profile for the country. 3.1.2 Technical characteristics of DPVs The solar installed capacity is assumed based on the information provided in Table 1. Bangladesh has a 166 million population by 2020, and the average household size (the number of people per household) is estimated to be 4.5 (World Bank, 2020). Electricity access in Bangladesh is 93% (BPDB, 2020a). This implies more than 34 million households have electricity access. If we assume about 3% of the total residential building with grid electricity access (or about 1 million households) adopt DPV, the total residential DPV capacity will be 1371 MW as calculated in Table 1. Note that these numbers are hypothetical and presented for illustration. The main analysis is independent of these assumptions. Similarly, we considered 5.75 million sq.m. commercial complex and 6 million sq.m. industrial complex for DPV installation in the commercial and 11 industrial sectors, respectively. The DPV installed capacities in these two sectors are 719 MW and 750 MW, respectively. Table 1. Deriving DPV size by sector Residential Commercial Industrial Households (HH) considered for DPV installation, Number 1,029,200 Area available for DPV in each HH on average, sq.m. 10.654 Size (area) of a typical residential DPV, sq.m. 2 Capacity of a residential DPV module, kW 0.25 Area considered for DPV for commercial complexes, sq.m. 5,750,000 Size (area) of a typical commercial DPV, sq.m. 2 Capacity of a commercial DPV module (kW) 0.25 Area considered for DPV in industrial complexes, (sq.m) 6,000,000 Standard size area of a typical Industrial DPV (Sq.m.) 2 Capacity of a standard industrial DPV module (kW) 0.25 Installed capacity (MW) 1,370.6 718.8 750.0 Note: Selection of space (areas) or the number of residential, commercial, and industrial premises are for illustration only. They do not represent the existing plans or policies of the government. The typical size of a DPV panel is 2 sq.m. and the standard power rating of a panel is 0.25 kW. Again, these parameters could change over time along with technological change. There is the loss of power from solar panels for many reasons, such as the overcast sky and design features (if the PV system includes batteries and inverters), air quality (e.g., dust). Some studies have estimated system losses. For example, Quansah and Adaramola (2018) the loss ranging from 18% to 39% in Ghana. Kumar et al. (2019) report 20% to 32% energy losses in PV systems in various locations in India as estimated by various studies. Since DPV does not need batteries and inverters as DPV owners export electricity to DICOMs anytime their generation exceeds their demand, we assume a 20% system loss annually. The DPVs installed in all three sectors (i.e., residential, commercial, and industrial) have 20 years of economic life. 12 3.1.3 Economic characteristics of DPVs Economic characteristics include capital costs, discount rate, and exchange rate. The capital costs of DPVs are US$840/kW, US$817/kW, and US$700/kW, respectively, for residential, commercial, and industrial applications (IRENA 2020). We increased and decreased the capital costs by 25% for a sensitivity analysis. We used 6%, 8% and 10% discount rates. The exchange rate is 80 taka (local currency) for one US dollar. 3.2 Characteristics of load and generation mix of the existing grid Electricity exchanges between the grid and DPV depend on the latter’ load profiles. However, we need to survey potential DPV owners to estimate their load profiles. The survey of residential, commercial, and industrial customers to determine their load profiles is beyond the scope of the study. Instead, we assume that DPV owners have the same load profiles as the grid or DISCOMs supplying electricity to these consumers. The system load profiles obtained from BPDB are presented in Table 2. 3.3 Other data Other data includes electricity tariff, system loss, Bangladesh Power Development Board’s annual financial statements. The latest electricity tariff was set by the Bangladesh Energy Regulatory Commission (BERC) on February 27, 2020, and it is available from BERC (2020). The net metering guidelines prepared by the Ministry of Power, Energy and Natural Resources (MPENR, 2018) sets the rule for net metering settlements. According to this rule, a distribution utility sends the customers (DPV owners) the electricity bill accounting for electricity it receives from the DPV owner in each billing period. If the DPV owner is a net exporter of electricity, the net exports will be carried over to the next billing period. At end of the fiscal year (June), the DPV owner receives payment from the utility for their annual net sales. The rate DPV owners get for their net sale to utility is equal to the utility’s tariff for high voltage (33 kV) customers, which varies between 7.57 to 11.28 taka per kWh (MPENR, 2018; BERC, 2028). Note that DPV owners are obliged to pay demand changes and other fixed charges for all billing periods (months) even if they are net sellers of electricity to the grid. 13 Table 2. Electricity system load profiles of Bangladesh 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 WD 0.66 0.67 0.66 0.64 0.69 0.68 0.67 0.65 0.64 0.62 0.64 0.65 0.64 0.63 0.61 0.65 0.64 0.65 0.67 0.66 0.65 0.66 0.66 0.64 Jan HD 0.53 0.62 0.68 0.67 0.67 0.65 0.62 0.65 0.62 0.58 0.53 0.49 0.60 0.65 0.65 0.64 0.62 0.57 0.53 0.61 0.66 0.66 0.63 0.53 WD 0.68 0.70 0.69 0.66 0.64 0.62 0.69 0.70 0.71 0.68 0.68 0.69 0.68 0.67 0.68 0.68 0.67 0.66 0.70 0.69 0.68 0.68 0.68 0.65 Feb HD 0.67 0.65 0.70 0.68 0.68 0.68 0.67 0.66 0.63 0.69 0.68 0.67 0.65 0.61 0.56 0.60 0.69 0.69 0.68 0.66 0.61 0.55 0.59 0.68 WD 0.68 0.69 0.68 0.76 0.78 0.81 0.80 0.81 0.78 0.76 0.78 0.77 0.79 0.81 0.82 0.79 0.77 0.75 0.70 0.65 0.69 0.67 0.69 0.68 Mar HD 0.65 0.69 0.72 0.71 0.70 0.71 0.69 0.68 0.79 0.82 0.81 0.78 0.77 0.77 0.74 0.80 0.82 0.81 0.78 0.76 0.72 0.69 0.71 0.63 WD 0.80 0.71 0.77 0.80 0.80 0.80 0.77 0.74 0.74 0.75 0.75 0.77 0.76 0.77 0.81 0.87 0.87 0.85 0.86 0.86 0.83 0.82 0.82 0.83 Apr HD 0.84 0.81 0.78 0.76 0.74 0.76 0.75 0.74 0.71 0.67 0.66 0.76 0.77 0.79 0.77 0.76 0.74 0.73 0.78 0.85 0.88 0.88 0.84 0.85 WD 0.94 0.92 0.90 0.93 0.92 0.94 0.94 0.95 0.91 0.89 0.86 0.85 0.91 0.94 0.95 0.96 0.97 0.97 0.97 0.96 0.94 0.92 0.91 0.90 May HD 0.93 0.93 0.94 0.95 0.97 0.97 0.96 0.95 0.93 0.93 0.93 0.92 0.89 0.78 0.89 0.94 0.94 0.94 0.92 0.91 0.90 0.89 0.91 0.93 WD 0.93 0.92 0.93 0.94 0.94 0.95 0.96 0.96 0.96 0.96 0.95 0.97 0.98 0.93 0.87 0.84 0.81 0.88 0.85 0.82 0.84 0.86 0.86 0.92 Jun HD 0.87 0.88 0.89 0.89 0.93 0.94 0.93 0.94 0.96 0.97 1.00 1.00 0.98 0.98 0.90 0.94 0.88 0.83 0.82 0.79 0.78 0.88 0.90 0.88 WD 0.86 0.85 0.84 0.85 0.85 0.87 0.86 0.86 0.88 0.86 0.85 0.85 0.87 0.88 0.87 0.87 0.86 0.83 0.81 0.82 0.83 0.81 0.82 0.81 HD 0.82 0.80 0.79 0.79 0.82 0.86 0.86 0.85 0.84 0.83 0.80 0.82 0.85 0.86 0.87 0.86 0.86 0.81 0.85 0.83 0.81 0.80 0.78 0.83 Jul WD 0.83 0.80 0.80 0.81 0.83 0.84 0.84 0.84 0.85 0.84 0.84 0.84 0.85 0.87 0.87 0.86 0.84 0.84 0.82 0.85 0.83 0.84 0.81 0.83 Aug HD 0.77 0.78 0.78 0.77 0.78 0.80 0.78 0.83 0.86 0.85 0.85 0.86 0.85 0.86 0.86 0.86 0.85 0.85 0.85 0.85 0.83 0.84 0.81 0.79 WD 0.88 0.89 0.88 0.88 0.89 0.90 0.88 0.87 0.87 0.86 0.84 0.85 0.84 0.84 0.85 0.87 0.85 0.83 0.82 0.86 0.87 0.89 0.89 0.89 Sep HD 0.90 0.88 0.89 0.88 0.87 0.79 0.88 0.88 0.87 0.86 0.85 0.84 0.82 0.86 0.86 0.86 0.85 0.83 0.84 0.83 0.82 0.81 0.78 0.89 WD 0.77 0.76 0.75 0.77 0.76 0.76 0.80 0.80 0.80 0.87 0.89 0.89 0.89 0.88 0.88 0.87 0.86 0.88 0.88 0.88 0.85 0.83 0.80 0.79 Oct HD 0.73 0.73 0.77 0.76 0.73 0.72 0.74 0.74 0.79 0.79 0.78 0.76 0.82 0.89 0.88 0.89 0.89 0.88 0.87 0.86 0.83 0.81 0.80 0.76 WD 0.72 0.69 0.67 0.68 0.68 0.67 0.68 0.68 0.67 0.67 0.67 0.67 0.65 0.65 0.68 0.68 0.66 0.74 0.75 0.75 0.76 0.77 0.74 0.73 Nov HD 0.73 0.70 0.65 0.61 0.68 0.68 0.69 0.67 0.65 0.60 0.59 0.65 0.65 0.64 0.61 0.71 0.73 0.73 0.72 0.69 0.65 0.64 0.74 0.75 WD 0.66 0.64 0.65 0.65 0.66 0.64 0.65 0.64 0.63 0.60 0.64 0.65 0.64 0.63 0.61 0.60 0.61 0.63 0.63 0.65 0.67 0.66 0.65 0.64 Dec HD 0.63 0.65 0.66 0.64 0.61 0.56 0.50 0.63 0.65 0.66 0.65 0.63 0.61 0.57 0.62 0.62 0.60 0.58 0.64 0.66 0.66 0.65 0.65 0.63 WD refers to workdays, HD refers to weekends and holidays. Load profile refers to a fraction of the load in all periods (hours) of the maximum (peak) load in a given year. Source: BPDB (2020b) 14 The transmission system loss is 2.9%, and the distribution system loss is 8.9% (BPDB, 2020a). We use the annual report of BPDB for FY2019-20 for a financial statement (BPDB, 2020a) to estimate the power generation costs of BPDB. The calculation accounts for the government’s direct subsidies and other operating expenses. If the government subsidies and other operating expenses are not accounted for, the cost of the BPDB power supply would be 6.7 taka per kWh (BPDB, 2020a). If the subsidy and other operating expenses are also accounted for, the cost of BPDB power would be 7.06 taka per kWh or 5.4% higher (BPDB, 2020a). The exchange rate used in the study is 80 taka per US dollar. 4. Results and Analysis The impacts of DPV would be different across the different stakeholders: DPV owners or investors; DISCOM; the national electricity utility; and the nation as a whole, excluding the environmental benefits of DPV and including environmental benefits. All results are discussed here with alternative assumptions or scenarios. 4.1 Impacts on DPV owners/investors 4.1.1. Impacts on electricity (energy) consumption Figure 2 shows the annual aggregated hourly load curves of DPV owners. If they are divided by 365, they can be interpreted as daily average hourly load curves. The load or hourly electricity demand is supplied through (i) DPV owners' production and (ii) utility or DICOM. If DPV owners' productions exceed their demand, they export it to the DISCOMs. The hourly representation of load is critical in the analysis of DPV because: (a) DPV owners demand changes across the hours, and (b) electricity generation from the installed DPV capacity varies across the hours. The highest generation occurs at 1 PM. During the day (8 AM to 5 PM), DPV owners’ annual generation would be higher than their demand; the excess generation is sold to DISCOM. As we discuss later, the capacity to meet their demand and export to the grid depends upon the 15 size of the DPV with respect to the own load of DPV owners. During the night, early morning, and evening time, DPV owners purchase electricity from DICOMs. The vertical height of the electricity exchange between the DPV owners and DISCOMs or the grid depends on the ratio of DPV owners’ peak load and installed capacity of their DPV (PL/IC ratio). The higher the PL/IC ratio, the lower would be capacity that the DPV owners offset their demand with their own generation or sell excess electricity to the grid. Figures 2(b) and 2(c) illustrate it. When the PL/IC ratio is 0.5 or the installed capacity of DPV is twice as high as DPV owners’ peak load, the DPV owners can meet 37% of their annual electricity demand through their own production (Figure 3). They can sell 30% of their annual generation to the grid. If DPV owners’ peak load is equal to the installed capacity of their DPV (PL/IC = 1), they can meet 25% of their annual demand through their production. Their sale to the grid drop to 3% of their total generation. When PK/PL ratio exceeds 1.5, DPV owners would not have excess electricity to sell to the grid. Figure 2: DPV owners’ annual load, self-consumption, and sales to the grid (GWh) (a) When PL/IC = 0.5) 450 400 350 300 250 GWh 200 150 100 50 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Hour DPV Owners's demand met through own production DPV Owners's demand met through utility electricity DPV electrcity sold to utility 16 (b) When PL/IC = 1) 900 800 700 600 500 GWh 400 300 200 100 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Hour DPV Owners's demand met through own production DPV Owners's demand met through utility electricity DPV electrcity sold to utility (c) When PL/IC = 2) 1,800 1,600 1,400 1,200 1,000 GWh 800 600 400 200 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Hour DPV Owners's demand met through own production DPV Owners's demand met through utility electricity DPV electrcity sold to utility Figure 3 provides the annual values for demand, generation, self-consumption, electricity exchanges with the grid. DPV’s annual electricity generation is 37% of its annual demand even 17 though its total installed capacity is twice as high as its peak load. This is because DPV does not produce electricity for almost half of the time (11 hours) in a day. If the DPV capacity is equal to the peak load of DPV owners, their annual generation would be 25% of their annual demand. For DPV owners, therefore, the economics of DPV depends on the relative size of the DPV with respect to its peak load and the timing of the peak load (during the day or evening). It will be shown later in this section that the amount of sale of DPV electricity to the grid and the corresponding price are the critical factors on the viability of DPV from the perspective of consumers (i.e., DPV owners). Figure 3: DPV owners’ annual electricity demand, generation, self-consumption, and electricity exchanges with the grid (GWh) 40,000 30,000 20,000 10,000 0 DPV Owners's total DPV Owners's total DPV Owners's DPV Owners's DPV electrcity sold demand generation demand met demand met to utility through own through utility production electricity PL/IC = 0.5 PL/IC = 1.0 PL/IC = 2 4.1.2 Impacts on electricity bills The DPV owners benefit from their DPV in two ways. First, they reduce their electricity bills for the electricity they buy from the grid, and secondly, they get additional revenues by selling the excess electricity to the grid. The net effect would be the reduction of their costs for electricity consumption. Table 3 provides a comparative picture of electricity costs for residential, commercial, and industrial consumers which install DPV. As can be seen from the table, the DPV scheme analyzed in this study reduces the total annual costs of DPV owners’ electricity 18 consumption by 25%, 29%% and 34% in residential, commercial, and industrial sectors. Overall, the net cost savings would be 29%. The DPV owners in the industrial sector would experience the highest savings because the grid electricity tariffs for the industrial customers are higher than those for residential and commercial customers. Table 3. DPV benefits: costs savings, revenue from electricity sales, and net benefits (Million Taka/Year) Costs and benefits item Residential Commercial Industrial Total Utility bills in the absence of DPV 33,416 21,485 24,764 79,664 Utility bills in the presence of DPV (excluding net metering 20,943 13,250 14,902 49,095 and cost of own production) Utility bill savings (excluding net metering and cost of own 12,473 8,235 9,862 30,570 production) (37%) (38%) (40%) (38%) Revenue from sales to the grid 5,566 2,919 3,046 11,531 Cost of own production from DPV 9,800 4,998 4,469 19,267 Net benefits (including net metering and cost of own 8,239 6,155 8,439 22,833 production) (25%) (29%) (34%) (29%) Note: Figures in parenthesis are % savings of electricity consumption expenditure due to the DPV The net benefits or net savings in the annual electricity bill are influenced by several factors considered in the analysis, including the price offered by the grid for the excess electricity of DPV owners, the relative size of DPV with respect to DPV owners’ demand, the cost of electricity production using DPV, the discount rate, etc. Below we discuss the effects of these factors on DPV owners’ net savings. Effects of net metering price: Figure 4 illustrates the effects of net metering price on DPV owners’ net reduction of their electricity bill. If the grid offers a net metering tariff that is equal to the production costs of DPV electricity, DPV owners’ net savings decrease significantly, 10 percentage points in the residential sector and 8 percentage points in the commercial sectors. This is because the production costs of DPV are 53%, 62% and 71% smaller than the offered net metering tariff in the residential, commercial, and industrial sectors, respectively. Because of the large difference in the offered net metering prices and the production costs of DPV, DPV owners reduce their electricity bills even if they are not paid for the excess electricity they export to the grid. However, their benefits would drop by 1.6 times in the residential sector to 3.1 times in the industrial sector. Effects of the relative size of DPV with respect to DPV owners' own load: The ratio of DPV owners’ peak load (PL) to their installed DPV capacity (IC) is a critical factor to determine DPV’s 19 benefits to the owners. A PL/IC ratio smaller than one indicates that the size of installed capacity of DPV is greater than DPV owners’ own load. It implies a higher potential for exporting excess electricity to the grid. On the contrary, if the PL/IC ratio is greater than 1, the export potential declines and becomes zero at a threshold depending on the solar energy profile in the location where the DPV is installed. Figure 5 illustrates how the DPV owners’ benefits (i.e., reduction of their electricity bills) diminish with an increasing PL/IC ratio. When the PL/IC ratio increases from 0.5 to 1.0, DPV owners' net savings decrease by about a half. When the PL/IC ratio is further increased to 2, DPV owners' savings drop by 75%. This result suggests that whether or not a DPV owner significantly benefits from DPV depends on how big the size of its DPV compared to its own electricity demand. Figure 4: Effects of net metering price on DPV owners’ savings on electricity costs (% of the electricity costs prior to DPV) 37% PL/IC=0.5 34% DR = 6% 29% 29% 29% 29% 25% 26% 22% 22% 20% 20% 14% 15% 14% 8% Current price offered Utility's sales price DPV production costs Not purchased by the utility Residential Commercial Industrial Total DPV owners in Bangladesh are still able to export electricity to the grid even if their DPV size is equal to their load. This is because peak load occurs in the evening; a DPV with equal size of DPV owner’s peak load would have higher generation potential than DPV owner’s demand during the day and therefore can export electricity to the grid. While this is true for the residential DPV owners and those commercial and industrial owners whose peak load occurs in the evening, this may not be the case for those industrial and commercial DPV owners whose peak loads occur during the day. However, since we do not have peak load information by type of DPV owners, 20 there might be some overestimation of electricity sold from commercial and industrial DPV owners to the utility. Figure 5. Effects of DPVsize with respect to DPV owners own load (% of the electricity costs prior to DPV) Offered price 34% DR = 6% 29% 29% 25% 19% 15% 15% 12% 10% 8% 7% 6% PL/IC = 0.5 PL/IC = 1 PL/IC = 2 Residential Commercial Industrial Total Effects of production costs of DPV: One of the reasons for the relative attractiveness of DPV is that the production costs of electricity from DPV are much lower than the electricity supply costs of the grid. We have checked if the economics of the DPV holds when the production costs of DPV increases. Production costs of DPV increase when the capital costs of the DPV increase or the discount rate used to annuitize the capital costs are increased. Figures 6a and 6b present the impacts on DPV owners' benefits (i.e., savings in their electricity costs) when the production costs change due to a change in capital costs or discount rates. If the capital costs are increased by 25%, DPV owners’ savings decrease by 30% in the residential sector, 20% in the commercial sector and 13% in the industrial sector. On the other hand, if the capital cost of DPV further decreases, say by 25%, DPV owners' savings increase by the same proportions. If the discount rate is increased from 6% (the value we use in the main analysis) to 8%, DPV owners' savings will decline by 21% in the residential sector, 14% in the commercial sector, and 9% in the industrial sector. A further increase in discount rate substantially reduces DPV owners' cost savings. If the discount rate increases to 10%, the benefits of DPV owners decrease by 43% in the residential sector, 29% in the commercial sector, and 19% in the industrial sector. 21 Figure 6: Effects of production costs of DPV on DPV owners electricity cost savings (% of the electricity costs prior to DPV) (a) Effects of capital costs PL/IC = 0.5 Offered price 39% 34% DR = 6% 34% 35% 32% 29% 29% 30% 25% 23% 23% 17% CAPEX CAPEX (+25%) CAPEX (-25%) Residential Commercial Industrial Total (b) Effects of discount rates PL/IC = 0.5 34% Offered price 31% 29% 29% 28% 25% 25% 24% 20% 20% 20% 14% 6% 8% 10% Residential Commercial Industrial Total 4.1.3 Benefits-to-costs ratio It would be more appropriate to compare the benefits with costs to measure the attractiveness of the DPV from DPV owners’ perspective. Figure 7 presents the benefit-to-cost (B/C) ratio under various conditions defined by the alternative values of input variables. In the 22 main analysis, the benefits to DPV owners would be almost two times as high as their costs in the residential sector. It would be further higher in the commercial and industrial sectors, with a benefit-to-cost ratio, 2.2 and 2.9. The reasons for variations of benefit to cost ratios under the alternative values for input variables are the same as discussed above. Figure 7: Benefit to cost ratios under alternative values for input variables (%) 10% Discount rate 8% 25% decrease CAPEX 25% Increase PL/IC =2 PL/IC ratio PL/IC =1 Zero Net metering price DPV production costs Utility's sales price Analysis Main 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 Total Industrial Commercial Residential 23 4.2 Impacts on the utility In Bangladesh, electricity is supplied to the retail customers through two channels. First, the national electric utility company (BPDB) sells electricity to its retail customers, and the second BPDP sells electricity to distribution companies, which are responsible for selling electricity to retail customers. There are six distribution companies. However, all these companies are state- owned companies. Therefore, we aggregated all these distribution companies and the BPDB’s distribution wing as a single utility for this study.5 Deployment of DPV affects the utility through multiple channels. While DPV causes the utility to lose its market as DPV owners use their own electricity to meet their load, the utility benefits through: (a) savings in transmission and distribution losses, (b) savings from electricity generation by the marginal plants. We also assume that the utility is mandated to buy surplus electricity from DPV owners. Although the price the utility pays to DPV owners for their surplus electricity could be different from the utility tariff to the consumers, we assume these prices are the same. This is because even if DPV owners get a higher price for the surplus electricity, they still have to pay for the distribution line services which are owned by the utility. Moreover, DPV owners still have to pay the demand charge imposed by the utility on them. Figure 8(a) presents the impacts of DPV on utility through various channels and also the net gain or loss when DPV is assumed to substitute average electricity generation from the system and fuel cost is subsidized. In this case, the utility will lose its markets with a value of almost 31 billion taka due to the DPV, which is about 6% of the revenue in the absence of DPV. At the same time, it does not need to produce electricity that would cost 9.4 billion taka. It also saves 4.7 billion taka through avoided transmission and distribution losses. Overall, the utility would suffer a loss of 16.6 billion taka or 3.2% of its revenue in the absence of DPV. 5 If we do not consider a single distribution utility, then we need to do the analysis for the seven distribution companies (six DISCOMs and BPDP’s distribution wing), it will make the calculation too cpmplicated and we need a large volume of data such as sales and revenues by customer types from each distribution entities. We need to carry out the entire analysis for each DISCOMs because of difference in electricity prices weighted by customers. We need to divide the DPVs across the DISCOMs for which we do not have a basis. 24 The reduction in revenue is sensitive to prices of fuel that would have been used to generate electricity in the absence of DPV. According to BPDB (2020), the system-wide average fuel cost of electricity generation was 1.64 taka per kWh in the 2019-2020 fiscal year. However, this cost represents a subsidized cost and generation replaced by the DPV is the average generation of the system. If the DPV replaces a generation at the margin and the fuel cost of the marginal plant follows the international price (i.e., opportunity cost), the utility’s loss or gain varies significantly. Using the international prices of fuels and standard heat rates for a given power generation technology, the fuel costs for power generation would be taka 2.14/kWh for natural gas combined cycle, taka 2.65/kWh for natural gas simple cycle, taka 3.31/kWh for natural gas steam turbine technology, taka 15.3/kWh for fuel oil-fired steam turbine technology and taka 21.34/kWh for diesel-fired internal combustion engine technology. The gains to the utility vary from -2.67% to 18.5%, depending upon the type of fuel used by the marginal power plant (Figure 8b). Figure 8. Impacts on electricity utility (a) Impacts on utility when DPV is assumed to substitute average electricity mix and average fuel cost is subsidized (Million Taka) 9,377 4,695 -16,599 -30,671 Utility's sales revenue Fuel cost savings T&D loss savings Net gain (+) or loss (-) 25 (b) Impacts on utility when DPV substitutes different types of generation at the margin and fuel costs are not subsidized (percentage change in the utility revenue as compared to the situation without DPV) 19.05% 11.97% -1.37% -1.00% -2.66% -2.11% -3.22% Average system Natural gas Natural gas Natural gas Coal steam Fuel oil steam Diese engine mix (subsidized combined cycle turbine (fuel steam turbine turbine (fuel turbine (fuel (fuel costs fuel costs (fuel costs costs (fuel costs costs costs Tk21.34/kWh) Tk1.64/kWh) Tk2.14/kWh) Tk2.65/kWh) Tk3.31/kWh) Tk3.65/kWh) Tk15.3/kWh) 4.3 Impacts at the national level The impacts of the DPV from the national perspective can be divided into two types: impacts without accounting for environmental or climate change benefits and impacts including the climate change benefits. In either case, the benefits include avoided fuel costs of electricity generation from the utility, avoided costs of transmission and distribution losses. Note that the benefits are sensitive to the type of generation the DPV replaces. Figure 9 presents the benefits of the DPV from the national perspective with and without the values climate change benefits. The benefits to the country from the DPV is 2.1% of the total electricity supply costs if we assume that that the DPV replaces all types of generation, on average, even if the value of carbon benefits is not accounted for. If the climate change benefit is accounted for using a carbon price of $40/tCO2, total benefits increases by 1.5 percentage point to 3.6%vof the total electricity supply costs. If the DPV replaces fule oil-based generation in the margin, the total benefits increase to 15.6% of the total electricity supply costs in Bangladesh. 26 Figure 9. Impacts of DPV from the national perspective (Benefits due to DPV as percentage of the total electricity supply costs in Bangladesh) 21.5% 19.0% 15.8% 13.6% 6.6% 5.1% 3.6% 4.2% 3.8% 3.5% 2.9% 3.5% 2.1% 2.5% System average Gas CC Gas Turbine Gas Steam Coal HFO Diesel Total benefits excluding climate change benefits Total benefits including climate change benefits 5. Conclusions and Policy Implications The distributed solar PV or DPV technology is getting the attention of investors/consumers and policy makers, especially in countries where governments have already introduced net metering tariffs or are planning to do so. Considering the declining costs of solar PV and also depending upon the solar profile in a given jurisdiction, a DPV system could be economically viable. However, the economic viability could vary across countries or jurisdictions depending upon several variables. This study develops a detailed economic model and applies it for an economic analysis of DPV in a developing country, Bangladesh. The model accounts for hourly solar generation profiles and hourly electricity load profiles for each month, thereby allowing electricity exchanges between the DPV owners and the electricity production and distribution utilities. Separate analyses are carried out for the residential, commercial, and industrial sectors because the costs of DPV and electricity tariffs vary across these sectors. The analysis is carried out from the perspectives of different stakeholders: DPV owners/investors, electricity utilities, and the nation as a whole (or the societal perspective). 27 The analysis finds that DPV systems are economically attractive to consumers or DPV owners or investors in all sectors. The net savings of DPV owners are 25% (residential sector) to 34% (industrial sector) of corresponding electricity bills in the absence of DPV. If the benefits are compared with costs of installation and operation of the DPV, the benefits are about two times as high as the costs in the residential sector and about three times as high as the costs in the industrial sector. The benefit-to-cost ratio of the commercial sector falls in between those of residential and industrial sectors. The benefits to DPV owners are sensitive to several input variables, including the relative size of the DPV capacity with respect to their peak load, the price offered to DPV electricity sold to the grid (or net metering price), and the cost of electricity production from the DPV, which is sensitive to the capital costs of DPV and discount rate. The higher the relative size of DPV with respect to their peak load, the larger would be the benefits as the DPV owners will get opportunities to export more electricity to the grid. The same is true if the net metering tariff is higher. The lower costs of DPV electricity through lower CAPEX and a lower discount rate increase the benefit-to-cost ratio. The nation as a whole benefits from the DPV even if environmental benefits are not accounted for. The total national gain depends on the type of generation DPV replaces. If it replaces the average fuel mix of the generation system, the gain would be 2% of the total electricity generation costs in the absence of DPV. On the other hand, if the DPV replaces the marginal generation, the benefit would much higher. For example, if the DPV replaces fuel oil-based generation, the gain would be 14% of the generation costs in the absence of the DPV. When accounted for environmental benefits, total national or societal gain would increase further. The study shows that total benefits, including climate change mitigation benefits, with the value of CO2 mitigation US$40/tCO2, would vary from 3.6% to 21.5% of the total generation costs depending upon the type of generation the DPV replaces. Whether the national utility gains or loses again depends on the type of generation the DPV replaces. If the DPV replaces fuel oil or diesel, which are marginal fuels during the day, the utility gains. On the other hand, if we assume that the DPV replaces all generations, the utility does not benefit from the DPV. It suffers a small loss of about 3% of its revenue in the absence of DPV. This analysis shows that DPV is clearly beneficial for the DPV owners and the country as a whole. Its benefits to the electricity utilities are conditional on the type of generation DPV 28 replaces. Even if the DPV does not make clear benefits to the utility in Bangladesh or any other countries where utilities are state-owned, DPV should be promoted because it benefits society as a whole. After all, the principal objective of a state-owned electricity utility is to provide reliable electricity to consumers at prices regulated by the government. In many countries, including Bangladesh, electric utilities are subsidized through direct and indirect subsidies. They enjoy indirect subsidy through cheaper or subsidized input fuel (e.g., natural gas in Bangladesh) and also direct budget subsidy from the government. If the DPV is beneficial from the national perspective, then the government could mandate the utilities to facilitate DPV by offering unconstrained access to DPV generated electricity and fair net metering pricing. The government could compensate the utilities’ loss, if there any, by reallocating the part of overall or national benefits, for example, the benefits coming through the sale of carbon credits. It can also be done through reallocation of DPV owners’ benefits by readjusting the net metering pricing. References Ahmad, Ali (2021). Distributed energy cost recovery for a fragile utility: The case of ´Electricit´e du Liban. Utilities Policy 68 (2021) 101138. Bangladesh Electrcity Regulatory Commission (BERC) (2020). Bangladesh Electrcity Regulatory Commission Notification. No. 28.01.0000. 012.04.003.20.652. Dated 27 February 2020. BERC. Dhaka, Bangladesh. Bangladesh Power Development Board (BPDB) (2020a). Annual Report 2019-20. BPDB, Dhaka, Bangladesh. Bangladesh Power Development Board (BPDB) (2020b). System Load Profile Data. BPDB, Dhaka, Bangladesh (obtained through personal communication). D.A. Quansah, M.S. Adaramola (2018). Ageing and degradation in solar PV modules installed in northern Ghana. Solar Energy, Vol. 173, pp. 834-847. Energy Sector Management Assistance Program (ESMAP) (2021). From Sun to Roof to Grid: Distributed PV to Empower Developing Countries. Unpublished Report. World Bank. Washington, DC. Energysage (2021). Net metering for home solar panels. https://www.energysage.com/solar/101/net-metering-for-home-solar-panels/?rc=seia. 29 Global Solar Atlas (GSA, 2020). Global Solar Atlas Database, Version 2.0, https://globalsolaratlas.info. Holdermann, Claudius Johannes Kissel and Jürgen Beigel (2014). Distributed photovoltaic generation in Brazil: An economic viability analysis of small-scale photovoltaic systems in the residential and commercial sectors. Energy Policy, Vol. 67, pp. 612-617. International Renewable Energy Agency (IRENA) (2020). Renewable Power Generation Costs in 2019. IRENA, Abu Dhabi. Ministry of Power, Energy and Natural Resources (MPENR) (2018). Net Metering Guidelines – 2018. MPENR, Dhaka, Bangladesh. Mitscher, M., R. Rüther (2012). Economic performance and policies for grid-connected residential solar photovoltaic systems in Brazil. Energy Policy, Vol. 49, pp. 688-694. Nallapaneni, Manoj Kumar, Ramjee Prasad Gupta Mobi Mathew, Arun kumar Jayakumar, Neeraj Kumar Singhe (2019). Performance, energy loss, and degradation prediction of roof- integrated crystalline solar PV system installed in Northern India. Case Studies in Thermal Engineering, Vol. 13: 100409. Qadrdan, Meysam, Nick Jenkins and Jianzhong Wu (2018). Smart Grid and Energy Storage. Soteris A. Kalogirou (ed.). McEvoy's Handbook of Photovoltaics: Fundamentals and Applications. Chapter 3, pp. 915-928. Academic Press. Timilsina, Govinda (2020). Demystifying the Costs of Electricity Generation Technologies. World Bank Policy Research Working Paper, WPS 9303. World Bank, Washington, DC. 2020. Vilaça Gomes, P., N. Knak Neto, L. Carvalho, J. Sumaili, J.T. Saraiva, B.H. Dias, V.Miranda, S.M. Souza (2018). Technical-economic analysis for the integration of PV systems in Brazil considering policy and regulatory issues. Energy Policy, Vol. 115, pp. 199-206. 30