Electrification and the Poverty Trap
Does an electricity connection help the very poorest households?
To many people, connecting rural households to the electricity grid in places like Kenya seems like a great way to help the world’s neediest. After all, the 1 billion people who currently live without electricity are some of the poorest people in the world. But, what if electricity connections mainly benefit the relatively wealthy in that group?
Development economists worry about what they call the “poverty trap,” a version of the idea that it takes money to make money. If that’s true at the lower end of the income distribution, a household needs capital to escape poverty — maybe the cash to buy a sewing machine to start a small tailoring business — which is exactly what very poor people don’t have. The effect can reinforce itself, meaning that poverty will persist from generation to generation.
An increasing body of research explores whether the poverty trap exists in practice. For example, people have shown that the stresses associated with living in poverty (hunger, poor sleep, etc.) can lead to lower cognitive skills and lower ability to do physical labor, both of which contribute to lower income levels. It’s also a lot harder to borrow money when you’re poor, which makes it hard to invest in capital to start a business. (This article summarizes recent literature on the topic, some of which supports and some of which refutes the importance of a poverty trap.)
In recent research on rural electrification, my co-authors Ken Lee, Ted Miguel, and I see an example of what could be a poverty trap. Here’s what’s going on:
Evidence of a Poverty Trap
To understand the impacts of rural electrification, we identified a set of 150 villages in two counties in Western Kenya. Then, we randomly selected half of those villages (75) to serve as a treatment group, while the other 75 served as a control group. In the treatment group, we again randomly divided the 75 villages into 3 sub-groups. We offered villagers in one of those sub-groups a free connection to the electricity grid, which would ordinarily cost a household about $400. Households in the other 50 villages got lower subsidies – $150 (meaning they had to pay $250) in one group and $250 (meaning they had to pay $150) in another. The control group could connect to the grid at the regular $400 price.
In the 25 villages that got a free connection, basically everyone accepted our offer and got connected. In the 50 villages that got a subsidy but still had to pay $150-250 dollars for the connection, only about 10-20% of the households got connected. One thing we saw, perhaps not surprisingly, is that the households that decided to get a connection in the 50 villages with the low subsidies were a lot better off economically than the households that got connected in the villages offered a free connection. They were more than twice as likely to have a formal bank account, their incomes were twice as high, they had 60% more chickens, etc.
After villagers had their electricity connections for a couple years, we went back and surveyed all of the households. Overall, we saw very few differences between the households that got an electricity connection and those that didn’t – incomes, health, assets, kids test scores, etc. were the same between our treatment and control groups. This was an intellectually fascinating but humanly depressing result, which we elaborated on in this post two years ago.
When we looked more closely at the data, though, we saw that the (relatively wealthier) households in the 50 villages that were offered a low connections subsidy seemed better able to do something with their electricity connection. They bought more appliances, they were more likely to start work outside of farming, they saved more money on kerosene, their household wealth was higher.
For example, the figure below reports the results for appliances and household wealth. The blue bar to the left of each figure reflects the control group, where almost no one got a new connection. The orange bar in the middle reflects the households in the communities that got a free connection. If anything, they had slightly fewer appliances and fewer assets. The green bar on the right shows that the (relatively wealthier) households in the low connection subsidy villages acquired on average 1.5 more appliances after they got a connection and their household wealth went up by about $500 relative to the control group.
Source: Lee, Miguel and Wolfram, Journal of Economic Perspectives, 2020
There weren’t very many households in the right-hand group in our overall sample since only a small share of households got a connection when they had to pay for it. This is why we didn’t see any impacts on average. This also means that the conclusions are a bit speculative (that is, they’re not all statistically significant).
Addressing the Poverty Trap to Benefit from Electrification
But, the patterns in the figure above are consistent with the poverty trap and bear further exploration. If someone comes and gives you a free electricity connection, you can’t start a business or get ahead economically unless you can also afford to buy appliances to use the electricity.
Our study does not prove that this is an example of the poverty trap. It could be, for example, that it was more profitable to be one of a few people to have a connection, which was the case in the 50 low-connection-subsidy villages. Maybe, for example, people in the low-subsidy group went into business providing cell phone charging to their neighbors, but that’s only a profitable business if very few of your neighbors have their own connection. This would be an example of correlation not causation. But, recent work by a pair of researchers at Duke University supports the notion that electrification is more valuable in areas with high incomes.
I do not mean to suggest that we should give up on electrification efforts for the very poorest households. There may be policy tweaks, such as connections accompanied with a free sewing machine or other appliance, to help households escape the poverty trap when they gain electricity. But, without further considering these possibilities, a broad brush application of a grid connection policy may not bear all the fruits that we expect.
Keep up with Energy Institute blogs, research, and events @energyathaas.
Suggested citation: Wolfram, Catherine. “Electrification and the Poverty Trap” Energy Institute Blog, UC Berkeley, February 18, 2020, https://energyathaas.wordpress.com/2020/02/18/electrification-and-the-poverty-trap/
Categories
Catherine Wolfram View All
Catherine Wolfram is the William F. Pounds Professor of Energy Economics at the MIT Sloan School of Management. She previously served as the Cora Jane Flood Professor of Business Administration at the Haas School of Business at UC Berkeley. From March 2021 to October 2022, she served as the Deputy Assistant Secretary for Climate and Energy Economics at the U.S. Treasury, while on leave from UC Berkeley. Before leaving for government service, she was the Program Director of the National Bureau of Economic Research’s Environment and Energy Economics Program, Faculty Affiliate of the Energy Institute at Haas from 2000 to 2023, as well as Faculty Director of the Energy Institute from 2009 to 2018. Before joining the faculty at UC Berkeley, she was an Assistant Professor of Economics at Harvard. Wolfram has published extensively on the economics of energy markets. Her work has analyzed rural electrification programs in the developing world, energy efficiency programs in the US, the effects of environmental regulation on energy markets and the impact of privatization and restructuring in the US and UK. She is currently working on several projects at the intersection of climate and trade. She received a PhD in Economics from MIT in 1996 and an AB from Harvard in 1989.
Energy Institute Blog 
great bit of research and although at first it is counter-intuitive it makes a lot of sense.
I read the article and the research paper with great interest. In India the fastest pace of rural electrification happened in the 1960s and 70s (prior to dedicated programs for household electrification) when the country increased supply of electricity to villages for pumping ground water to increase food production. The data for “pump set energisation” was routinely presented in India’s five year plans. This was the so called green revolution in India which had two unintended externalities. One was the depletion of ground water (negative) and the other was rural household electrification (positive). The electrification programs of the 80s and 90s depended on grid extensions during the green revolution.
Congratulations on those articles, Catherine! I wonder if those households are really “trapped” in the usual sense of non-convexities (e.g. fixed costs, increasing rates of return to capital) or they just have low endowments ala Schutz.
Is the welfare economics of rural electrification well developed? It looks rather weak to my less than expert eye. The typical claims of economies of scale to not apply to extensive (spatial) development, i.e. when you’re extending the system to lower density areas (Vernon Smith, 1079). Nor is rural electrification clearly justified on vertical equity grounds; the subsidies go disproportionately to better-off households. Finally, rural electrification is said to be justified on the grounds of “geographical equity,” a special case of horizontal equity. This might be true to the extent that location is random and exogenous, but clearly that is not the case. Low-endowment folks tend to locate in low-rent locations, including the cost of utilities. Surely there are more cost-effective redistributive instruments available, e.g. cash-contingent transfers.
Studies on the benefits of electrification also typically include a rather amorphous reference to economic growth and development, but the kind of network externalities that justify subsidies are concentrated in industry.
Willingness-to-pay estimates for rural electrification tend to give high values to TV, but where are the behavioral economists when you need them? TV-watching can stunt intellectual development, with typical recommendations being no screen time for children two years or less and not more than two hours for others.
Two obvious thoughts come to mind.
1: Statistically – what would be more interesting would be a comparison of people who started off at the same socio-economic level, since the article already says that the people who got connected in the control & low-subsidy were higher income, which I bet skews the data far more than anything else.
2: The absolute cost – my previous company (Lumeter) was the technology provider to many of the PAYG companies, and the median price of a system was $200, (paid off at approx $2/wk for 2 years). That $2/wk represented an ability to pay for the median person who did not already have electricity. The Kenyan connection price of $400 plus the cost of electriciity is going to in almost all cases cost significantly above that $2/wk ability to pay (it would take 4 years, just to pay the cost of connection, without any actual power).
Our observation with customers providing both minigrid and stand-alone systems were that there were about 10x more households connected stand-alone, and that the real customer of mini-grid and grid-electrification projects was the donors and aid agencies, which is why they generated so much more publicity (including research studies) than the people selling small stand-alone systems whose customer was the rural poor.
“Our observation with customers providing both minigrid and stand-alone systems were that there were about 10x more households connected stand-alone…”
mitra42, re: the stand-alone systems you encountered in rural Kenya – was the source of electricity renewables, diesel generators, or a mixture of the two?
“When we looked more closely at the data, though, we saw that the (relatively wealthier) households in the 50 villages that were offered a low connections subsidy seemed better able to do something with their electricity connection.”
Catherine, the reason may be due more to psychology than poverty. I think when most customers are offered a product for free they tend to assign little value to it, and may consider the same product more valuable after it’s been priced (and may even consider buying it).
Also, wouldn’t people who had invested in an electricity connection be more motivated to make good use of it – so they wouldn’t feel they had wasted their money?
I think that there is more to this story than is given in the article. In Tanzania, if one has electricity, one must prepay for the energy you use. Kenya and other African countries the model is the same. One has to go to a store and prepay. You are given a number that you put into the meter and the electricity comes on. When you have used the power you have paid for, the power is turned off at the meter. Everything in Africa is prepaid, including cell phone service. In fact, cell phones minutes are used as a form of money is rural areas of Kenya and Tanzania, because they can be transferred from one phone to another. If one connects the power for free and the people receiving the free connection can’t afford to buy the power there is little gained from the free connection. In poor households, the electricity is used for lighting, because it is cheaper than kerosene. The author is correct that being able to afford to buy the appliances is a factor in wealth generation. An interesting thing about Africa is that there are more households with cell phones than have installed electricity. It is also true in Africa that well meaning gifts such as clothes can put the people who sew clothes out of business, thus cutting off a source of income for many families.