4.2.2. Energy Access and Usage-Patterns

To gain a better understanding of the end-user adaptation to utilising different energy sources and the conditions of how to facilitate the productive use of energy to generate a stable demand and potentially positive economic impacts in the community, different energy access situations, energy consumption patterns and paymen<sup>t</sup> models have been compared. This evaluation is guided by two questions that are central to understand local

adaption and the impact of decentralised energy systems. Firstly, what type of energy sources are currently most used by rural communities? Secondly, how does the source of energy influences energy usage patterns?

Figure 4 illustrates respondents with different connection types in Uganda and Zambia and points towards higher access rates to the national grid, Solar Home Systems (SHS) and MGs in Uganda. While this is partially due to the selection process of respondents as for Uganda, also respondents with access to the national grid have been selected during the survey rounds, the data illustrates a clear tendency to higher energy access rates in rural Uganda compared to Zambia which correlates with World Bank data which presents a rural electrification rate of around 10% for Zambia compared to 38% in Uganda [60]. This data, however, is mainly focused on connections to the national grid. Around two-thirds of the respondents in Uganda have access to MGs, the national grid or SHS, in Zambia however, this figure stands at only around 20% with 80% of the Zambian respondents originally claiming to have no electricity access at all. A further survey round that focused on rural community members in Southern Zambia revealed, however, that despite a high share of respondents originally claiming to have no electricity access at all, granular data from two communities showed that over 80% of the respondents use some kind of energy sources like portable solar panels or car batteries.

**Figure 4.** Type of energy supply, Uganda and Zambia (Responses in %; N = 1037).

In both countries, the most prevalent energy source for rural households is SHS with 'Pay-as-you-go' (PAYG) or Rent-to-Own (RTO) purchase models to be seemingly more widespread in Uganda while one-off purchases seem to be more common in rural Zambia which is also due to ye<sup>t</sup> limited availability of these solutions which have been acknowledged as the most favoured type of financing and electricity access model for the Zambian respondents.

With regard to the second question focusing on energy usage pattern it could be hypothesised, that the ownership of appliances and the utilisation of the energy provided, including its productive use, are potentially diversifying and growing in the course of enhanced connectivity. The data presented in Figures 5 and 6 partially supports this hypothesis to some extent for MG and SHS users in both countries but the data also illustrates the ye<sup>t</sup> limited productive use of energy across all connection levels despite the fact that over 80% of the respondents are self-employed entrepreneurs or farmers. The national-level data indicates similar energy usage patterns for Uganda and Zambia with energy being mostly used for lighting, communication including charging devices and entertainment.

**Figure 5.** Ownership of appliances, Uganda and Zambia (Responses in %; N = 1037).

**Figure 6.** Energy usage patterns; Consumer Groups in Uganda and Zambia (Responses in %; N = 1037).

The availability of MG systems does not automatically trigger the possession of a wider array of appliances among the respondents such as fridges and the productive use of energy but reveals visible national differences instead. While Ugandan MG users have a higher but still limited tendency towards productive energy, this is of less significance to Zambian MG users which could be due to higher income and energy affordability levels as well as better access to appliances in Uganda. Users of the Ugandan Kalangala mini-grid for example pay between 0.18 and 0.22 \$ per kWh as of 2019 [70] while Zambian mini-grid users in Sinda pay around 0.26 \$ per kWh with income levels in rural Zambia being less than half compared to Uganda as illustrated earlier.

### 4.2.3. Energy Demands, 'Willingness to Pay' (WTP) and Productive Uses

The majority of the respondents in both countries indicated, that their current energy source is limited and not sufficient to meet their energy needs for example with regard to lighting, productive use or the connection of additional appliances. Consequently, the data presented in Figure 7 suggests a general desire of the respondents to upgrade their current energy supply especially in Zambia, where 100% of respondents indicated their willingness to upgrade while this figure is less pronounced in Uganda, especially among MG users. This might indicate, that the Ugandan MG has the tendency to better fulfil the energy demands of its customers while the Sinda MG in Zambia does not meet the energy- needs of the customers to large extent. Data from Southern Zambia also sugges<sup>t</sup> that consumers would not automatically prefer an MG connection over an SHS as around 55% of the respondents in Southern Zambia stated their preference of an SHS based on a rent-to-own model, independently from their current source of electricity. This dataset also indicated limited popularity of grid-connection provided by the national utility (Zambia Electricity Supply Corporation—ZESCO) as none of the respondents preferred that connection type which might partially be due to the high amount of national power-cuts of up to 12 h daily Zambia experienced in 2018/2019 due to a serious drought in the hydropower-dependent country. The expression of consumer preferences however does not eradicate the question of the impact of national grid-arrival in a market offering subsidised energy tariffs for on-grid connections versus the requirement of cost-reflectiveness in the off-grid sector.

**Figure 7.** Desire for upgraded electricity supply Consumer Groups in Uganda and Zambia (Responses in %; N = 1037).

The desire for an upgraded energy supply is largely mirrored by the willingness to pay for better energy access which is high in both countries as illustrated in Figure 8. With regard to the intended use of an upgraded energy connection, the responses show a wide array of preferences with significant variations among consumer groups with regard to cooking and cooling. Lighting, charging devices and productive uses and to some extent entertainment, however, are desired by the majority of all respondents in each consumer group as Figure 9 shows which indicates a ye<sup>t</sup> untapped consumer demand in many areas.

**Figure 8.** Willingness to pay for an upgraded connection, Energy-use demands, Consumer Groups in Uganda and Zambia (Responses in %; multiple responses; N = 1037).

**Figure 9.** Energy-use demands, Consumer Groups in Uganda and Zambia (Responses in %; multiple responses; N = 1037).

As the productive use of energy is still limited among all consumer groups in both countries, the majority of all respondents confirmed their demand to use energy more productively. Figure 10 illustrates the diversity of the forms of desired productive energy use between various consumer groups and communities. While the data suggests demandtrends for cooling, lights and charging of devices, the demands in other areas are much more diverse between various groups of respondents. Despite these variations, respondents expressed a strong demand for improved access to productive appliances including financing mechanisms that would be based on 'Rent-to-Own' or 'Pay-Go' models which were the preferred solution over one-off purchases which appears ye<sup>t</sup> to be a challenge.

**Figure 10.** Type of desired productive uses in Uganda and Zambia (Responses in %; multiple responses; N = 1037).

A direct comparison of two rural communities in Western Uganda and Southern Zambia that share similar features like energy access via SHS, low and intermittent income and the absence of wage labour level included presented in Figure 10 reveals that the demands for productive energy use vary despite their similarities. The data suggests that manufacturing and crafts activities are underrepresented, especially in Uganda and the clear trend towards using energy for service provision in both countries. Since most services depend on income patterns of 'their' customers, the data implicate the opportunity of training, capacity building and appliance financing mechanisms coupled with MG development to diversify local business structures to create local added value and positively impact the income situation in local communities.

### 4.2.4. Impacts of Electrification on Consumer Finance

Since low and intermittent income levels in the rural communities surveyed establish substantial challenges for the financial MG sustainability, the question for impacts of energy availability in the communities arises and more specifically whether energy availability did directly contribute to local income generation to stabilise MG cashflow. Although the overall majority of respondents reported the general improvement of life quality after getting a connection, either through SHS or MG as shown in Figure 11, the data on income change does not draw a conclusive picture with this regard.

**Figure 11.** General perceived community impact of electrification on communities in Uganda and Zambia among various subgroups; Responses in %.

As Figure 2 showed earlier for both countries, spending levels between MG users versus respondents without energy connection are generally higher with greater significance in Zambia where spending levels of MG users are almost double as high as those of non-connected respondents. For SHS users this correlation is less visible as SHS users in Uganda reported similar spending levels compared to non-connected respondents in Uganda while the spending levels for SHS users in Zambia are around 40% higher compared to non-connected respondents. Since the data presented in Figure 2 does not provide a blanket indication of energy access automatically generating higher incomes or whether merely consumers who already have a higher income can afford an MG or SHS connection, respondents have been specifically asked for the impact of getting electrified on their income. The responses presented in Figure 12 clearly indicate a positive impact on the income situation for MG and SHS users in Uganda while interestingly about two-thirds of the Sinda MG users in Zambia stated, that their income remained the same or has even decreased after getting electrified. This response pattern seems to replicate for SHS users in Zambia and is largely mirrored by the number of respondents reporting difficulties in energy paymen<sup>t</sup> among SHS users and to a larger extend MG users in Zambia as shown in Figure 13.

The impact of electrification on income, however, becomes inconsistent when respondents have been asked about their ability to pay for school fees before and after electrification. While a clear majority of MG users in Uganda and around 53% of SHS users in Zambia confirmed an increased ability to pay school fees after getting electrified, the majority of SHS users in Uganda and MG users in Zambia stated no or negative impact of electrification on the ability to pay for school fees.

**Figure 12.** Impact of electrification on income among various subgroups; Responses in %.

**Figure 13.** Difficulty in paymen<sup>t</sup> for electricity among various consumer subgroups; Responses in %.

Another indicator for the financial community impact of electrification is the development of business activity in the community. The response patterns with this regard reflect positive income development to some extend with the biggest perceived growth of business activity in the community after electrification for MG and SHS users in Uganda, to some extent among SHS users in Zambia but with no perceived positive impact among the respondents from the MG users in Zambia.

A clear majority of respondents across all consumer groups reported increased studying time of children after getting connected which indicates a positive impact on education in both countries.

These variations among the different indicators related to electrification and enduser income illustrate the complexity of this dimension and highlight the importance of a multidimensional approach of measuring the financial impact of energy systems on community level through a systematic evaluation of various indicators as a sole focus on household income when measure impact might not reflect the complex reality of rural communities. For exhaustive coverage, it would also require a measurement system that includes the non-direct monetary value components such as the ability to purchase certain goods, increase of months with income, ability to generate savings etc. which could be a result of changes in end-user income due to electrification.
