Urban demand for cooking fuels in two major African 2 cities and implications for policy

Abstract

78 electricity are too expensive to use for cooking purposes (2), or issues related to the unreliability of 79 electricity supply in many LMIC cities. Poor distribution networks and small-scale production also 80 mean that more affordable efficient biomass stoves can be difficult to find in many urban centers. As 81 such, household use of polluting fuels continues to be a major contributor to the cocktail of sources 82 that are increasingly making ambient air in urban areas unbreathable (3).

83
84 This paper focuses on urban cooking fuel demand in two of East Africa's largest and fastest growing 85 cities -Nairobi, Kenya and Dar es Salaam, Tanzania. Our double-bounded, dichotomous choice 86 contingent valuation (CV) experiment help us to understand how urban households would respond to 87 changes in the price of their preferred (main) cooking fuels (4). It acknowledges that households 88 typically stack fuels, and that households may react to higher prices by either maintaining their current 140 polluting fuels), and in 2020 we interviewed a representative sample of 1,100 households (largely 141 residing in informal settlements) across Dar es Salaam (SI Appendix, Figure S3). A detailed summary 142 of the characteristics of sampled households in both locations is presented in Table S1. Here, we 143 specifically describe the energy profile of the sampled households. In the Nairobi sample, LPG is the 144 most common cooking fuel (54%), followed by kerosene (29%) and charcoal (11%) (Panel B, Table 145 S1). In Dar es Salaam, charcoal is the most common cooking fuel (61%), followed by LPG (32%) 146 and kerosene (4%). In Nairobi, fuel procurement times for all major cooking fuels are similar, ranging 147 from 10-14 minutes per purchase. Average reported daily fuel collection times in Dar es Salaam are 148 much higher than in Nairobi, with firewood collection time being the highest (44 minutes per trip) 149 and kerosene collection time being the lowest (24 minutes). All Nairobi households in the sample 150 have electricity, and electricity access is slightly lower in the Dar es Salaam (85%) sample.
151 Households in the Nairobi sample are of slightly higher socio-economic status than those in the Dar 152 es Salaam sample. Per capita monthly expenses are 94 USD in Nairobi, compared to 78 USD in the 153 Dar es Salaam sample, and roughly 51% of the Nairobi sample had completed secondary schooling, 154 compared to 33% in Dar es Salaam.

155
156 Primary cooking fuel choices in response to price increases 157 For each of the three main cooking fuels in Nairobi (charcoal, kerosene and LPG) and Dar es Salaam 158 (firewood, charcoal and LPG), we assessed demand over randomly-assigned price increases that 159 ranged from 25% to 200% (SI Appendix, Table S2 presents the full range of prices offered for the 160 three fuels in both cities). The derived demand curves from responses to initial bids show a mostly 161 linear relationship between WTP for cooking fuel and price (Figure 1). This manuscript is a preprint and has not been peer reviewed. The copyright holder has made the manuscript available under a Creative Commons Attribution 4.0 International (CC BY) license and consented to have it forwarded to EarthArXiv for public posting.

162
163 Note: This figure shows the percentage of households in the Dar es Salaam and Nairobi samples reporting that they 164 would continue using their primary cooking fuel when faced with a given initial price increase. Price increases are 165 randomly assigned across survey respondents and baseline prices are converted into USD for comparison in this figure. 166 The initial bids in the price increases and willingness to maintain use responses are presented here.
167 Figure 1. Demand graph for cooking fuels in Nairobi and Dar es Salaam (initial bids only) 168 169 Among respondents in Nairobi that use charcoal as their primary cooking fuel, nearly half the 170 respondents were willing to maintain their primary fuel use under the lowest price increase of 25%, 171 but only 10% were willing to pay 200% more for the fuel. The WTP probabilities among primary 172 kerosene-using respondents were similar: ranging from 62% to 16% for these initial lowest and 173 highest bids, respectively. For primary LPG fuel respondents, the range of WTP probabilities was 174 somewhat higher, dropping from 93% to 30%. Regression analyses further show that the price 175 elasticities of maintaining use of each of these primary fuels are somewhat similar, ranging from -0.5 176 to -0.7 (Table 1, columns 1, 4, and 7). Yet, controlling for fuel stacking (i.e., use of multiple cooking 177 fuels), which represents an important strategy for coping with high fuel costs or unreliable fuel 178 alternatives, adds important nuance to these findings. Specifically, we find that accounting for 179 stacking leads to higher estimates of the price elasticity for charcoal and LPG (columns 2 and 8), 180 while kerosene users appear less price sensitive (column 5). Primary LPG-using households are 13 This manuscript is a preprint and has not been peer reviewed. The copyright holder has made the manuscript available under a Creative Commons Attribution 4.0 International (CC BY) license and consented to have it forwarded to EarthArXiv for public posting. license EarthArXiv 181 percentage points more likely to switch away from that fuel when they already use other cooking 182 fuels (column 8). Moreover, the higher the proportion of LPG consumed in total cooking fuel use, 183 the more likely they maintain this option (column 9). In SI Appendix, Section S3, we further explore 184 the determinants of cooking fuel stacking.   Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1 Household-level controls included are log of monthly per capita total household expenditures (in USD), household size, age, education and gender of household head, dependency ratio, whether household has saved money anywhere in the past year, whether household is connected to electricity, time taken to acquire firewood, charcoal, kerosene and LPG. As the sample for the number of primary firewood users in Dar es Salaam is small (22 households), we do not show results for that sub-sample.
212 increase and maintained primary use of their preferred fuel, we analyze the extent of the cooking 213 reduction they predicted they would make. In Nairobi (Table S5), we find that a 1 USD increase in 214 LPG price per L would reduce cooking with LPG by 9-12 percentage points. In Dar es Salaam (Table   215 S6), we find that a 1 USD increase in charcoal per kg and LPG price per L faced by primary users of 216 those fuels would reduce cooking with those fuels by 25-30 percentage points, and 8 percentage 217 points, respectively.

225
226 In Nairobi, the measures range from 0.14-0.2 USD/kg charcoal, 1.1-1.4 USD/liter kerosene, and 2.3-227 2.9 USD/kg LPG ( This manuscript is a preprint and has not been peer reviewed. The copyright holder has made the manuscript available under a Creative Commons Attribution 4.0 International (CC BY) license and consented to have it forwarded to EarthArXiv for public posting. license EarthArXiv This manuscript is a preprint and has not been peer reviewed. The copyright holder has made the manuscript available under a Creative Commons Attribution 4.0 International (CC BY) license and consented to have it forwarded to EarthArXiv for public posting. license EarthArXiv 256 how much relative charcoal prices would need to change to induce more substantial switching 257 towards clean fuels.

258
259 Distributional aspects of price change policies 260 It is critical to also understand the distributional impacts that fuel pricing policies might have across 261 the income distribution. Low-income households are likely to respond differently to price increases 262 than high-income households, especially when the switching costs are high. There is also a risk that 263 price increases on some fuels could be regressive.
264 To explore such aspects, we restrict the Dar es Salaam sample to households who cook mainly with 265 charcoal, given that over 60% households (n=672) use charcoal as their primary cooking fuel. In 266 Nairobi, we restrict the sample to households who cook mainly with kerosene, who constitute 30% 267 of the sample (n=104). These two subsamples represent households that policy makers might target 268 for transitioning to cleaner fuels. We then divide the income distribution into relatively low-and high- 272 In Dar es Salaam, high-income charcoal users are more likely than low-income charcoal users to 273 switch up the energy ladder to LPG for any given price increase, especially when the price increase 274 is large (Figure 2). Lower-income households are less likely to switch away from charcoal, but those 275 switching more often move to kerosene and firewood. Thus, large charcoal price increases in Dar es 276 Salaam at this time could be regressive, in the sense that low-income households either maintain the 277 use of charcoal, or switch down the energy ladder. In Nairobi, in contrast, where policies are more 278 supportive of clean options, low-income households appear more likely to switch to LPG than high-279 income households. High-income kerosene-using households are instead more likely to switch to This manuscript is a preprint and has not been peer reviewed. The copyright holder has made the manuscript available under a Creative Commons Attribution 4.0 International (CC BY) license and consented to have it forwarded to EarthArXiv for public posting. license EarthArXiv 280 charcoal, suggesting that these households' current preference for polluting fuel may reflect a 281 resistance to using LPG that should be further explored.

315
316 Our findings show that fuel switching patterns in response to price changes vary across the income 317 distribution and depending on the specific context, i.e., across these two East African cities. In both 318 locations, the price elasticity of demand for cooking energy overall is similar, and LPG demand 319 appears to be somewhat less price elastic than charcoal demand. WTP for LPG among respondents 320 cooking primarily with that fuel is also significantly higher than prevailing market prices. In relative 321 terms, however, the ratio of the market price to WTP is low for LPG compared to charcoal and 322 kerosene in Nairobi, while it is only low for LPG compared to firewood (and not compared to charcoal) 323 in Dar es Salaam. In the latter city, low-income charcoal users thus appear more entrenched in their 324 choice of cooking fuel and less likely to switch to cleaner LPG. Important for interpreting these 325 findings is the fact that the WTP estimates apply specifically to the subsamples using particular 326 primary fuels, rather than the overall sample in each city.
This manuscript is a preprint and has not been peer reviewed. The copyright holder has made the manuscript available under a Creative Commons Attribution 4.0 International (CC BY) license and consented to have it forwarded to EarthArXiv for public posting. license EarthArXiv

328
The extent to which different policy tools can be effective also depends crucially on the readiness of 329 the supply side to meet increased demand. Non-price instruments such as bans on charcoal that 330 effectively increase the price of charcoal in low enforcement contexts, or taxes, may be successful in 331 urban areas where a market for alternative fuels exists, but can be regressive when households have 332 limited access to affordable clean fuel alternatives, and may also induce back-sliding to even dirtier 333 fuels. LPG subsidies, while critical for fostering uptake, can also be regressive unless they are well 334 targeted to reach the poor. As such, other complementary mechanisms are essential, such as 335 supporting access to clean fuels by reducing upfront stove and canister investments, aiding the private 336 sector in developing efficient supply networks for fuel refills, and shifting preferences away from 337 polluting fuels with information and behavior change efforts.

338
This manuscript is a preprint and has not been peer reviewed. The copyright holder has made the manuscript available under a Creative Commons Attribution 4.0 International (CC BY) license and consented to have it forwarded to EarthArXiv for public posting.   Table S2 shows the randomized price 387 increases offered to households in both locations. If respondents responded positively to the initial 388 bid, they received a follow up question with a payment option that was double the initial bid; if 389 respondents replied in the negative to the initial bid, they received a follow up question of a payment This manuscript is a preprint and has not been peer reviewed. The copyright holder has made the manuscript available under a Creative Commons Attribution 4.0 International (CC BY) license and consented to have it forwarded to EarthArXiv for public posting. license EarthArXiv 390 option that was half the initial bid. In addition, respondents who declined to switch their main cooking 391 fuels were asked whether, and to what extent, they would reduce their cooking in response to the 392 price increase. This design allows us to assess both intensive and extensive margin responses to 393 hypothetical fuel price changes.

413
This manuscript is a preprint and has not been peer reviewed. The copyright holder has made the manuscript available under a Creative Commons Attribution 4.0 International (CC BY) license and consented to have it forwarded to EarthArXiv for public posting. license EarthArXiv