Rural Electrification in Africa - Lee et al (2020) Flashcards
one of the biggest issues with electricity infrastructure in SSA
reliability
- few experience regular and reliable power
about half the population isn’t connected or they are connected and it doesn’t work
electrification randomisation
working with Kenya Power, the national electricity utility, where the project randomised subsidies for grid connections to rural Kenyan households
government trying to boost electrification rates since only about 5% of households were connected to the grid
looking at results on household impacts after 18 months and again after 32 months
living under grid
grid expansion can exploit existing investments
by 2016, over 3/4 of Kenyans lived under grid (<1km from a low voltage line but are unconnected themselves)
centre for global development estimates that tens of millions of Nigerians were living under grid, with 100M Africa-wide (Leo et al 2014)
household rural electrification demand in Kenya at baseline
5% of households connected in all groups
household rural electrification demand in Kenya after the program
when it was free, almost all households get connected
- some where they didn’t want to fill out forms or elderly who didn’t want the connection/resources wasted on them
polar extreme with no subsidy
- some households got connected but a relatively small share
at the intermediate subsidy levels, kind of in between
- the cheaper the price, the more people get connected
- downwards-sloping demand curve
demand and cost curves
demand for electricity connections falls sharply at higher prices
- even prices far below the cost of connection
costs fall at higher coverage (higher levels of connected households) implying large economies fo scale
- projects have a fixed upfront cost but falling marginal costs
demand is much lower than the cost curve and even the MC curve
- not a socially desirable investment
- if demand reflects household valuation of this good, society should not be spending an order of magnitude of money more than what people are willing to pay from it
pre-analysis plans
details the main research hypotheses, outcome variables and regressions
prevents researcher from focusing on a subset of findings, rather than more representative patterns in the data
- prevents data mining, etc.
may help resist pressure from other scholars/policymakers/donors to report certain findings and suppress others
publicly-registered plans can help build a culture of greater transparency and openness in research
electrification impacts at 18, 32 months
households that received electricity connections show an immediate increase in electricity consumption compared to control
economic outcomes (employment, earnings, hours worked, assets or consumption) do not show significant gains
- most treatment effects are 0, total hours worked go down a bit, etc.
non-economic outcomes also have 0 effect (no gains in political knowledge, self-reported health, perceived security or test scores)
exception of the significant gain in self-reported life satisfaction at 0.16sd
- very significant but not a huge effect
interpretation to increased life satisfaction
maybe more freedom/flexibility in the day
connection to electricity in rural Kenya means you’re a big deal
- people get connected and feel like they have something others don’t
- asset as a status symbol
maybe missing other aspects of social interaction and other benefits in terms of time use
concern of experimenter demand effects
- people say what they think the researcher wants to hear
why didn’t electrification have a huge impact?
are they too poor to do anything meaningful with electricity?
households with connections saw an increase in electricity consumption but only by roughly $2 monthly
- small amount relative to <5kWH monthly (enough to power a light bulb and charge a cell phone)
households consume less than 25kWh per month
- without appliances, electricity doesn’t do much for you
number of household appliances increased from 2 to 2.3 in high subsidy group
- barely increases
setting where there can be extended blackouts
combination of this program and one that subsidises appliances might have gone together better
welfare implications of a mass rural electrification program
no meaningful gains in household economic or educational outcomes after up to 32 months
- similarly no evidence of meaningful impacts up to 5 years later
bottom line that electrification may have a high return for industry or urban households but other investments may yield higher benefits for rural households (health, education, agriculture, etc.)
